WO2017113370A1 - Voiceprint detection method and apparatus - Google Patents

Abstract

A voiceprint detection method and apparatus. The method comprises: when a matching degree between a voiceprint feature of an audio signal portion and a pre-set voiceprint feature exceeds a pre-set threshold value, and a matching degree between an exhaled airflow feature of a determination signal portion and an exhaled airflow feature of the audio signal portion exceeds the pre-set threshold value, determining that a voiceprint detection result indicates that the detection has succeeded (S206). By means of the voiceprint detection method and apparatus, the security of voiceprint unlocking is improved.

Description

Method and device for detecting voiceprint Technical field

The present invention relates to the field of electronic technologies, and in particular, to a method and apparatus for voiceprint detection.

Background technique

With the development of mobile terminals and intelligent interactions, terminal devices have become an indispensable part of people's daily lives. In order to protect the private information stored by the user in the terminal device, most terminal devices are provided with a password protection unlock function. When the terminal device is in the locked state, the user can only unlock the terminal device by inputting the correct password. At present, there are many methods for unlocking a terminal. Since voice unlocking has higher security than other unlocking methods, it has become a widely used unlocking method. The terminal device or application software provides a function of unlocking voice, verifying the user through voice unlocking, further unlocking the terminal device, or providing services.

At present, the voice unlocking mainly authenticates the user through the voiceprint unlocking, and compares the sound signal input by the user with the preset sound signal when unlocking, and if it is determined that the voiceprint input by the user matches the preset voiceprint, it is determined to be a person. , then unlock it.

However, with the current voiceprint unlocking method, the recording attack cannot be prevented, that is, the text recognized by the user is recorded, and the voice-recognition text of the recording is played out to unlock the voiceprint, and the voiceprint can be successfully unlocked. The soundprint unlocking has a safety hazard and the safety is not high.

Summary of the invention

The invention provides a method and a device for detecting a voiceprint, which improves the security of voiceprint unlocking.

In a first aspect, the method for detecting voiceprint according to the present invention comprises: detecting whether a sound signal is present by a terminal, and if the terminal detects a sound signal, the terminal receives the sound signal, and the terminal extracts the audio signal portion and the judgment signal portion of the sound signal, The voiceprint feature of the audio signal portion is compared with the preset voiceprint feature, and the expiratory airflow characteristic of the judgment signal portion is compared with the exhalation airflow characteristic of the audio signal portion, and the voiceprint feature of the audio signal portion is preset. When the matching degree of the voiceprint feature exceeds a preset threshold, and the degree of matching between the expiratory airflow characteristic of the signal portion and the expiratory airflow characteristic of the audio feature portion exceeds a preset threshold, it is determined that the voiceprint detection result is successful. Make the terminal recognize the sound When the signal is used, the sound signal is divided into an audio signal portion and a judgment signal portion, thereby realizing double recognition of the sound signal, and at the same time, effectively avoiding the situation that the user blows the mouth while playing the recording, and improves the security of the voiceprint unlocking.

In a possible implementation manner, the feature of the expiratory airflow that is greater than a preset airflow threshold in the portion of the determination signal is received; the characteristics of the expiratory airflow are quantized; and the quantized expiratory airflow characteristic corresponds to the text corresponding to the audio signal portion. Comparing the characteristics of the expiratory airflow; if the matching of the characterized expiratory airflow characteristic with the expiratory airflow characteristic of the audio signal portion exceeds a preset threshold, determining the expiratory airflow characteristic of the signal portion and the audio characteristic portion The matching of the expiratory airflow characteristics exceeds a preset threshold. By comparing the quantified expiratory airflow characteristics with the expiratory airflow characteristics of the audio signal portion, whether the matching degree of the quantified expiratory airflow characteristics and the expiratory airflow characteristic of the audio signal portion exceeds a preset threshold is determined. Improve the accuracy of blow signal recognition.

In a possible implementation manner, the expiratory airflow characteristic is compared with a preset airflow threshold, and if the expiratory airflow characteristic is greater than the preset airflow threshold, the expiratory airflow feature is quantized to 1; otherwise, The expiratory flow characteristic is quantized to 0; if at least one of the following two conditions: the expiratory airflow characteristic is quantized to 1, and the text corresponding to the audio signal portion is an aspirated sound; the expiratory airflow characteristic is quantized to 0, and the audio signal portion is The corresponding text is a non-aspirate sound, and the matched expiratory airflow characteristic and the expiratory airflow characteristic of the audio signal portion exceed a preset threshold. The characteristics of the expiratory flow are quantified by comparing the characteristics of the expiratory flow with the preset airflow threshold.

In a possible implementation manner, it is determined whether the pointing direction feature of the determining signal portion and the pointing direction feature of the audio signal portion are within a preset range; when the matching relationship between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and determining that the matching degree of the expiratory airflow characteristic of the signal portion and the expiratory airflow characteristic of the audio characteristic portion exceeds a preset threshold, and determining a pointing direction characteristic of the signal portion and a pointing direction characteristic of the audio signal portion are in advance When the range is set, it is judged that the voiceprint detection result is successful. By judging whether the audio signal and the air blowing signal in the sound signal are from the same directivity direction, the situation that the playing recording and the mouth-type blowing direction may be inconsistent is effectively avoided, and the safety of the voiceprint unlocking is improved.

In a possible implementation manner, the angle of the pointing direction of the determining signal portion and the pointing direction of the audio signal portion are respectively compared with the preset pointing angle threshold; if the angle of the pointing direction of the signal portion and the audio signal portion are determined The angle of the pointing direction is smaller than the preset pointing angle threshold, and then the pointing direction characteristic of the signal portion and the pointing direction characteristic of the audio signal portion are preset. Within the scope. By comparing the angle of the pointing direction of the judgment signal portion and the angle of the pointing direction of the audio signal portion with the preset pointing angle threshold, respectively, whether the pointing direction characteristic of the judgment signal portion and the pointing direction characteristic of the audio signal portion are in advance Set within the scope.

In a possible implementation manner, the sensing temperature characteristic of the determining signal portion is compared with a preset temperature threshold; when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and determining The matching degree of the expiratory airflow characteristic of the signal part and the expiratory airflow characteristic of the audio characteristic part exceeds a preset threshold value, and the pointing direction characteristic of the signal part and the pointing direction characteristic of the audio signal part are within a preset range, and the judgment signal part is When the perceived temperature characteristic is greater than or equal to the preset temperature threshold, it is determined that the voiceprint detection result is successful. By determining whether the perceived temperature characteristic of the signal portion is greater than or equal to the preset temperature threshold, it is determined that the sound signal received by the terminal is from the user, not the recorded electronic device, thereby avoiding the recording attack and improving the security of the voiceprint unlocking.

In a possible implementation manner, before the terminal extracts the audio signal portion and the determination signal portion of the sound signal, the method further includes: the terminal separating the sound signal into the audio signal portion and the determining signal portion; specifically, the terminal adopts the sound signal a filter of a preset frequency is filtered to obtain an audio signal portion; the terminal filters the sound signal by using a filter of a second preset frequency to obtain a judgment signal portion; wherein the filter of the first preset frequency is a high-pass filter The filter of the second preset frequency is a low pass filter. The sound signal is separated into an audio signal portion and a judgment signal portion by passing the sound signal through a filter of a preset frequency.

In a possible implementation manner, the voiceprint feature of the audio signal portion includes: at least one of a voiceprint waveform and a signal frequency; and at least one of the following two cases: the voiceprint waveform of the audio signal portion and the pre- Aligning the characteristic waveforms of the voiceprint samples; comparing the signal frequency of the audio signal portion with the characteristic frequency of the preset voiceprint sample; if the matching between the voiceprint waveform of the audio signal portion and the preset voiceprint sample characteristic waveform exceeds The threshold value is set; and/or, the matching between the signal frequency of the audio signal portion and the characteristic frequency of the preset voiceprint sample exceeds a preset threshold, and the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds Set the threshold. Comparing the voiceprint waveform of the audio signal portion with the preset voiceprint sample feature waveform; and/or comparing the signal frequency of the audio signal portion with the preset voiceprint sample feature frequency to implement the audio signal portion The voiceprint feature is compared to the preset voiceprint feature.

In a possible implementation manner, the method further includes: collecting, by the terminal, a sound signal sent by the user, performing feature analysis on the sound signal, acquiring a preset voiceprint feature, and storing the sound signal. Collect users in advance through the terminal The sound signal is emitted, and the sound signal is analyzed as a preset voiceprint feature and stored to ensure the accuracy of the preset voiceprint feature, thereby improving the accuracy of matching the voiceprint feature of the audio signal portion with the preset voiceprint feature. Sex, which improves the security of voiceprint unlocking.

In a possible implementation manner, the method further includes: acquiring, by the terminal, an airflow feature that is exhaled when the user outputs a sound corresponding to the sound signal. The acquisition of the expiratory airflow characteristic of the judging signal portion is performed to ensure that the expiratory airflow characteristic of the judging signal portion is compared with the expiratory airflow characteristic of the audio signal portion.

In a possible implementation manner, the method further includes: acquiring, by the terminal, a direction of the user outputting a sound corresponding to the sound signal. The obtaining of the pointing direction feature of the judging signal portion is performed to ensure that the pointing direction feature of the judging signal portion and the pointing direction feature of the audio signal portion are within a preset range.

In a possible implementation manner, the method further includes: acquiring, by the terminal, a temperature when the user outputs a sound corresponding to the sound signal. The acquisition of the sensing temperature characteristic of the determination signal portion is performed to ensure that the sensing temperature characteristic of the determination signal portion is compared with the preset temperature threshold.

In a second aspect, the terminal provided by the present invention includes: a detecting module, configured to detect whether a sound signal is present; a receiving module, configured to receive a sound signal; and an extracting module, configured to extract an audio signal portion and a determining signal portion of the sound signal; a matching module for comparing the voiceprint feature of the audio signal portion with the preset voiceprint feature; comparing the expiratory airflow characteristic of the determination signal portion with the expiratory airflow characteristic of the audio signal portion; wherein, exhaling The airflow characteristic is a feature of the airflow exhaled when the user outputs the sound corresponding to the sound signal; and the determining module is configured to: when the matching degree of the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and determine the signal portion When the matching degree of the expiratory airflow characteristic and the expiratory airflow characteristic of the audio feature part exceeds a preset threshold, it is determined that the voiceprint detection result is the detection success. When the terminal recognizes the sound signal, the sound signal is divided into the audio signal portion and the judgment signal portion, thereby realizing the double recognition of the sound signal, and at the same time, effectively avoiding the user blowing the air while playing the recording, and improving the voiceprint unlocking. Security.

In a third aspect, the terminal provided by the present invention includes: a microphone and a processor; a microphone for detecting whether there is a sound signal; if detecting a sound signal, receiving a sound signal; and a processor for extracting an audio signal portion of the sound signal And judging the signal portion; comparing the voiceprint feature of the audio signal portion with the preset voiceprint feature; comparing the expiratory airflow characteristic of the determination signal portion with the expiratory airflow characteristic of the audio signal portion; wherein, the expiratory airflow The feature is a feature of the airflow exhaled when the user outputs the sound corresponding to the sound signal; when the matching degree of the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the expiratory airflow characteristic and the audio of the signal portion are judged Exhalation of the characteristic part When the matching degree of the airflow feature exceeds a preset threshold, it is determined that the voiceprint detection result is successful. When the terminal recognizes the sound signal, the sound signal is divided into the audio signal portion and the judgment signal portion, thereby realizing the double recognition of the sound signal, and at the same time, effectively avoiding the user blowing the air while playing the recording, and improving the voiceprint unlocking. Security.

In a fourth aspect, the present invention provides a non-transitory computer readable storage medium storing computer instructions for causing an apparatus for controlling a cache to perform an operation in the above method.

The method and device for detecting voiceprint according to the present invention detect whether there is a sound signal through a terminal, and if the terminal detects a sound signal, the terminal receives the sound signal, and the terminal extracts the audio signal portion and the judgment signal portion of the sound signal, and the audio signal portion The voiceprint feature is compared with the preset voiceprint feature, and the expiratory airflow characteristic of the judgment signal portion is compared with the exhalation airflow characteristic of the audio signal portion, and the voiceprint feature and the preset voiceprint feature of the audio signal portion are compared. When the matching degree exceeds the preset threshold, and the matching degree of the expiratory airflow characteristic of the signal part and the expiratory airflow characteristic of the audio characteristic part exceeds a preset threshold, it is determined that the voiceprint detection result is the detection success. When the terminal recognizes the sound signal, the sound signal is divided into the audio signal portion and the judgment signal portion, thereby realizing the double recognition of the sound signal, and at the same time, effectively avoiding the user blowing the air while playing the recording, and improving the voiceprint unlocking. Security.

DRAWINGS

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below. Obviously, the drawings in the following description are only It is a certain embodiment of the present invention, and other drawings can be obtained from those skilled in the art without any inventive labor.

1A is a schematic diagram of a scene for unlocking a voiceprint according to an embodiment of the present invention;

FIG. 1B is a schematic diagram of a scenario for setting a voiceprint password according to an embodiment of the present invention; FIG.

2 is a flowchart of a method for detecting voiceprint according to Embodiment 1 of the present invention;

3A is a schematic diagram of quantification of a blowing signal according to Embodiment 1 of the present invention;

3B is a schematic diagram of quantification of a blowing signal according to Embodiment 2 of the present invention;

4 is a schematic diagram of a process of voiceprint detection according to Embodiment 1 of the present invention;

FIG. 5 is a flowchart of a method for voiceprint detection according to Embodiment 2 of the present invention; FIG.

6 is a schematic diagram of an angle of a pointing direction of a sound signal according to Embodiment 1 of the present invention;

7 is a flowchart of a method for voiceprint detection according to Embodiment 3 of the present invention;

FIG. 8 is a flowchart of a method for voiceprint detection according to Embodiment 4 of the present invention; FIG.

FIG. 9 is a schematic structural diagram of a terminal according to Embodiment 1 of the present invention;

10 is a schematic structural diagram of a terminal according to Embodiment 2 of the present invention;

FIG. 11 is a schematic structural diagram of a terminal according to Embodiment 3 of the present invention;

FIG. 12 is a schematic structural diagram of a terminal according to Embodiment 4 of the present invention;

FIG. 13 is a schematic structural diagram of a device for detecting a voiceprint according to Embodiment 1 of the present invention.

detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

FIG. 1A is a schematic diagram of a scene of a voiceprint unlocking according to an embodiment of the present invention. As shown in FIG. 1A, the terminal device or the application software provides a function of unlocking the voiceprint. The user verifies the user by unlocking the voiceprint by speaking the corresponding voiceprint password, further unlocking the device, or providing a service. Voiceprint recognition generally includes two types: 1. The text content recognized during voiceprint recognition is preset: each time the unlock is unlocked, the same user-preset text recognition is repeated (for example, sesame opens the door); or When the voiceprint is recognized, the electronic device randomly generates some text or digital passwords, and the user reads out the prompt random password to ensure the security of the voiceprint recognition; 2. The text content recognized during the voiceprint recognition is randomly set: setting When the user randomly says something, the electronic device recognizes the user after extracting the characteristic parameters of the user, and when the voiceprint recognition is required, the user can recognize the player as the speaker. FIG. 1B is a schematic diagram of a scene of a voiceprint password setting according to an embodiment of the present invention. As shown in FIG. 1B, a user can set a voiceprint password, and a voiceprint password can be defined in advance, for example, by defining a voiceprint password, the user After the voiceprint password "Sesame Opens" is spoken, the terminal successfully enters the user's voiceprint password through the microphone, and the user logs in the account through the voiceprint password, and the terminal determines whether to let the user log in to the account by verifying the voiceprint password input by the user.

FIG. 2 is a flowchart of a method for voiceprint detection according to Embodiment 1 of the present invention. As shown in FIG. 2, the method provided by the embodiment of the present invention includes:

S201: The terminal detects whether there is a sound signal.

It should be noted that the terminal according to the embodiment of the present invention has the function of receiving voice, and the terminal may include, but is not limited to, a mobile communication device such as a mobile phone or a tablet computer.

Specifically, when the user needs to unlock the verification, the user sends a sound signal (speech signal) to the terminal. For example, the voice signal sent by the user may be that the user speaks the preset voiceprint password “open sesame” or the user calls. The name of the voice assistant, such as "small ice", "hello google", etc., may also be a text or digital password randomly generated by the user to read the terminal, or the user may randomly say a paragraph. When the terminal is in the unlocked state, it detects whether there is a voice signal sent by the user. If the terminal detects the voice signal sent by the user when the terminal is in the unlocked state, that is, when the voiceprint recognition signal is detected, the voice signal sent by the user is recognized.

Optionally, in the embodiment of the present invention, the terminal is not always in the living voiceprint recognition mode, but when the terminal detects the voiceprint recognition signal, after entering the living voiceprint recognition mode, the voice signal sent by the user is identified. The terminal is in an unlocked (standby) state. When voiceprint recognition is required, the voiceprint recognition mode is entered, for example, the terminal enters the screen to be locked, the application software is to be unlocked, the user's mouth is close to the microphone, or the user is identified. When a microphone or other scene or a combination of any scene is combined, the live voiceprint recognition mode is entered. The mouth proximity microphone can be judged by a sensor such as a proximity sensor, an ultrasonic sensor, an infrared sensor, or the like, and enters the living voiceprint recognition mode. The living voiceprint recognition mode requires the terminal to open the corresponding module to perform corresponding analysis and processing on the received voiceprint identification signal, including, for example, a recording module, a voiceprint recognition module, a thermometer module, a light sensor module, a directional monitor module, Any module or combination of modules in the ultrasonic sensor and infrared sensor to enter the live voiceprint recognition mode. Optionally, the terminal in the embodiment of the present invention may also be in the living voiceprint recognition mode. When the voiceprint recognition signal is detected, the voice signal sent by the user may be identified. The implementation of the present invention mainly describes how to enter the living voiceprint recognition mode when the terminal detects the sound signal, but is not limited thereto.

It should be noted that the voiceprint is a sound wave spectrum of a sound signal (speech signal) displayed by an electroacoustic instrument. Due to different habits of different people's voices, different people's vocal airflow is different, resulting in sound quality and tone. There are differences, and each voiceprint is different. Voiceprint recognition is a type of biometric recognition to confirm whether a certain voice is spoken by a designated person. The voiceprint recognition signal is a sound signal (speech signal) detected when the terminal is in an unlocked state, and the voiceprint recognition signal includes a voice signal of the user voiceprint, and the terminal can recognize the sound signal according to the voiceprint recognition signal. Whether the user's voiceprint is the voiceprint of the specified user to confirm whether the detected voice signal is what the specified user said.

S202: If the terminal detects that there is a sound signal, the terminal receives the sound signal.

Specifically, when detecting a sound signal, the terminal can receive the sound signal through the microphone. Optionally, the terminal receives the sound signal and stores the received sound signal. In order to avoid the sound signal of the voiceprint recognition signal detected by the fault, the terminal may always be in the listening state, and buffer the received sound signal, so that when the terminal enters the living voiceprint recognition mode, the voiceprint identification signal is complete for analysis and processing. .

S203: The terminal extracts an audio signal portion and a judgment signal portion of the sound signal.

Specifically, the sound signal may include an audio signal of the user's speaking voice and a perceived temperature when the user speaks, and may also include an audio signal and a sound signal direction of the user's speaking voice, and may also include an audio signal of the user's speaking voice and the user speaking. The exhaled signal, the terminal may divide the sound signal into an audio signal portion and a judgment signal portion, wherein the audio signal portion may include a voiceprint feature of the audio signal in the sound signal, and the determination signal portion may include a perceived temperature when the user speaks, At least one of a direction of the sound signal and a signal of exhalation when the user speaks, for example, the terminal can obtain the perceived temperature of the voice signal in the voice signal by the terminal; the terminal can also obtain the directivity direction of the sound signal through the microphone array; The terminal can also obtain a signal of exhalation when the user speaks through a filter of a preset frequency (low pass filter).

S204: Align the voiceprint feature of the audio signal portion with the preset voiceprint feature.

Specifically, the terminal compares the voiceprint feature of the audio signal portion with the preset voiceprint feature, and determines whether the voiceprint feature of the audio signal portion matches the preset voiceprint feature.

In the embodiment of the present invention, before the terminal enters the standby state, the user may set the living voiceprint recognition in the terminal, including receiving a voice signal preset by the user, for example, giving the word “sesame opening”, the user reads the a preset text, the terminal records a user voice signal, the voice signal includes an audio signal that the user reads the preset text, the audio signal has a voiceprint recognition feature, and the voiceprint recognition feature of the audio signal is used as a preset sound Pattern features.

Optionally, the voiceprint feature may include at least one of a voiceprint waveform of the audio signal and a signal frequency of the audio signal. Comparing the voiceprint feature of the audio signal portion with the preset voiceprint feature may be by at least one of two things:

A case where the voiceprint waveform of the audio signal portion is compared with the preset voiceprint sample feature waveform Correct.

Another case: the signal frequency of the audio signal portion is compared with the preset voiceprint sample characteristic frequency.

S205: Compare the characteristics of the expiratory airflow of the judging signal portion with the characteristics of the expiratory airflow of the audio signal portion.

The exhalation airflow is characterized by the airflow exhaled when the user corresponding to the sound signal outputs the sound.

In the embodiment of the present invention, the terminal captures the microphone input by detecting a sound signal received by the microphone, using a tape recorder or the like. When people are pronounced normally, including the strength of the airflow and whether the vocal cords vibrate, when the vocalization, the exhaled airflow must open the glottis. Due to the Bernoulli effect, the glottis is returned, and the pressure under the glottis is large enough to open the glottis repeatedly. The opening and closing forms a periodic tremor, so there is an airflow exhalation when the sound is pronounced, which is referred to herein as a blowing signal, that is, the blowing signal is an outgoing airflow characteristic corresponding to the sound output by the user. For example, for example, when the user says "open", since the open word is a gas-sounding sound, when the user makes a "open" sound, the user needs to exhale the airflow to open the door to emit the sound, which is necessary to bring out some The airflow, the microphone receives the effective airflow as a blow signal. Since the frequency of the audio in the sound signal is about 300-3000 Hertz (Hz), the air blowing sound to the microphone is mainly a low-frequency signal, so the high-frequency component that is not blown can be filtered out by low-pass filtering to obtain the air blowing. Signal, thereby separating the audio signal and the blow signal.

Specifically, when the terminal detects that there is a sound signal, and the extracted judgment signal portion includes a blow signal, the terminal converts the audio signal into a corresponding text, and determines that the exhalation airflow characteristic of each word or word of the text is a breath sound or The air-free tone is compared with the expiratory airflow characteristic of the audio signal portion to determine whether the user's air blowing signal matches the audio signal. For example, when the user recognizes a sample in a preset voiceprint, the voice is sounded for a certain word, but when the voiceprint is verified, the voice is unvoiced when the word is pronounced, and it is determined that the user's blow signal does not match the audio signal.

It should be noted that, according to the voice habits of different users, the terminal may learn the characteristics of the exhalation airflow of each user's audio signal from at least one of the user's daily call and the voice assistant according to the user's voice habit, for example, When the user speaks a particular word or word, the blow is larger, while the same word or other user blows less, to improve the accuracy of the user's expiratory flow characteristics.

S206: When the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and determining the characteristics of the expiratory airflow characteristic of the signal portion and the expiratory airflow characteristic of the audio feature portion When the degree of distribution exceeds the preset threshold, it is determined that the voiceprint detection result is successful.

Specifically, the terminal compares the voiceprint feature of the audio signal portion with the preset voiceprint feature, determines that the voiceprint feature of the audio signal portion matches the preset voiceprint feature, and the terminal determines the expiratory airflow characteristic of the signal portion and The characteristics of the expiratory airflow in the audio signal portion are compared, and the characteristics of the expiratory airflow of the signal portion are matched with the characteristics of the expiratory airflow of the audio feature portion. The voiceprint detection result is successful. At this time, the terminal is unlocked, and the user can be at the terminal. Complete the corresponding operations, such as unlocking the phone, logging in to WeChat, and so on.

For example, the terminal may first determine whether the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, if the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature does not exceed The preset threshold value does not match the voiceprint feature of the audio signal portion and the preset voiceprint feature, and the terminal determines that the voiceprint detection fails, and the terminal can directly exit the voiceprint detection mode; if the voiceprint portion of the audio signal portion and the preset sound If the matching degree of the pattern feature exceeds a preset threshold, it is further determined whether the matching degree of the expiratory airflow characteristic of the determination signal portion and the expiratory airflow characteristic of the audio characteristic portion exceeds a preset threshold value, and if the expiratory airflow characteristic of the signal portion is determined The matching degree of the expiratory airflow characteristic with the audio feature portion exceeds a preset threshold, and the voiceprint feature of the audio signal portion matches the preset voiceprint feature, and the exhalation airflow characteristic of the signal portion and the exhalation of the audio feature portion are judged. The airflow feature is matched, the terminal determines that the voiceprint detection is successful, and the terminal unlocks; if the signal part of the expiratory airflow characteristic and the audio feature is judged Matching feature of the expiratory flow does not exceed a preset threshold, wherein the expiratory flow expiratory flow characteristic signal portion and an audio portion is determined characteristic does not match, the terminal determines a failure detector voiceprint, voiceprint detection mode terminal exits.

It should be noted that the preset threshold may be determined according to actual conditions. For example, if the voiceprint feature matching accuracy in the terminal is high, the preset threshold may be set to 95%. If the voiceprint feature matching accuracy in the terminal is low, Set the preset threshold to 90%.

The voiceprint detection method provided by the embodiment of the present invention detects whether there is a sound signal through the terminal. If the terminal detects a sound signal, the terminal receives the sound signal, and the terminal extracts the audio signal portion and the judgment signal portion of the sound signal, and the audio signal portion The voiceprint feature is compared with the preset voiceprint feature, and the expiratory airflow characteristic of the judgment signal portion is compared with the exhalation airflow characteristic of the audio signal portion, and the voiceprint feature and the preset voiceprint feature of the audio signal portion are compared. When the matching degree exceeds the preset threshold, and the matching degree of the expiratory airflow characteristic of the signal part and the expiratory airflow characteristic of the audio characteristic part exceeds a preset threshold, it is determined that the voiceprint detection result is successful, so that the terminal recognizes the sound. Signal, The sound signal is divided into an audio signal part and a judgment signal part to realize double recognition of the sound signal, and at the same time, the user can effectively avoid the situation that the user blows the mouth while playing the recording, and the safety of the voiceprint unlocking is improved.

Further, in the embodiment shown in FIG. 2, the method for voiceprint detection further includes:

An expiratory airflow characteristic that is greater than a preset airflow threshold in the portion of the determination signal is received.

The characteristics of the expiratory flow are quantified.

The quantified expiratory airflow characteristics are compared to the expiratory airflow characteristics of the audio signal portion.

The matching degree between the expiratory airflow characteristic of the signal portion and the expiratory airflow characteristic of the audio characteristic portion exceeds a preset threshold, including:

The quantified expiratory airflow characteristic matches the expiratory airflow characteristic of the audio signal portion exceeding a preset threshold.

In the embodiment of the present invention, after extracting the audio signal portion of the sound signal and the determining signal portion, the terminal determines whether the size of the expiratory airflow of the inflating signal is greater than a preset airflow threshold, and the receiving the determining signal portion is greater than the preset airflow threshold. The expiratory flow is quantified based on the size of the expiratory flow. The preset airflow threshold in the embodiment of the present invention may take 0.10 liters/second (L/s).

Further, in the embodiment shown in Figure 2, the characteristics of the expiratory flow are quantified, including:

The expiratory airflow characteristic is compared with a preset airflow threshold. If the expiratory airflow characteristic is greater than the preset airflow threshold, the expiratory airflow characteristic is quantized to 1; otherwise, the expiratory flow characteristic is quantized to zero.

The quantified expiratory airflow characteristic and the expiratory airflow characteristic matching degree of the audio signal portion exceed a preset threshold, and includes: at least one of the following two cases.

In one case, the expiratory airflow characteristic is quantized to 1, and the text corresponding to the audio signal portion is an aspirated sound.

In another case, the expiratory airflow characteristic is quantized to 0, and the text corresponding to the audio signal portion is an unspised sound.

In the embodiment of the present invention, for the quantification of the insufflation signal, the insufflation may be divided into several levels, for example, 10 levels. When the gas of the received insufflation signal is greater than or equal to the fifth level, the determination is met. The preset threshold determines that the blow signal is 1. When the gas of the received blow signal is less than the fifth level, it is determined that the preset threshold is not reached, and then the blow signal is determined to be zero. By quantifying the blow signal, the accuracy of the blow signal recognition can be improved. FIG. 3A is a schematic diagram of quantification of a blowing signal according to Embodiment 1 of the present invention. As shown in FIG. 3A, when the gas of the blowing signal reaches three levels, Then, it is judged that the blowing signal is 0. FIG. 3B is a schematic diagram of quantification of a blowing signal according to Embodiment 2 of the present invention. As shown in FIG. 3B, when the gas of the blowing signal reaches 8 levels, it is determined that the blowing signal is 1.

4 is a schematic diagram of a process of voiceprint detection according to Embodiment 1 of the present invention. As shown in FIG. 4, for example, a user sends a voiceprint identification signal “Opening a door of sesame”, and after the microphone of the terminal is received, the separation module separates the voiceprint recognition. The signal is an audio signal and a blow signal, and the audio signal is further sent to the voiceprint recognition module to complete the voiceprint recognition. After the voiceprint recognition, the audio-to-text module converts the audio into corresponding text, and determines each word of the text or The blow signal corresponding to the word is an aspirated sound or a non-aspirate sound. The blowing module quantifies the received blowing signal, defines 1 for the blowing signal equal to or greater than the threshold, 0 for the threshold, outputs the binary signal of the blowing signal, and judges the word output by the module to the audio to text module. Or the word is compared with the binary signal output by the blowing signal recognition module. For example, the blowing signal of the "sesame door opening" that the user says is "0", "0", "1", "0". For example, the user says that the "top" blow signal is "1" "1", and the user says that the "sport" blow signal is "0" "1".

FIG. 5 is a flowchart of a method for voiceprint detection according to Embodiment 2 of the present invention. Another specific implementation manner of the method provided by the embodiment of the present invention, as shown in FIG. 5, the method provided by the embodiment of the present invention includes:

S501: The terminal detects whether there is a sound signal.

S502: If the terminal detects that there is a sound signal, the terminal receives the sound signal.

S503: The terminal extracts an audio signal portion and a judgment signal portion of the sound signal.

S504: Align the voiceprint feature of the audio signal portion with the preset voiceprint feature.

It should be noted that the implementations of S501, S502, S503, and S504 are the same as those of S201, S202, S203, and S204. For details, refer to the descriptions of S201, S202, S203, and S204, and details are not described herein again.

S505: It is judged whether the pointing direction feature of the judgment signal portion and the pointing direction feature of the audio signal portion are within a preset range.

Specifically, the determining signal portion may include a pointing direction feature, wherein the pointing direction feature is a direction in which the user corresponding to the sound signal outputs the sound. In practical applications, the sound signal received by the terminal may have problems in that the audio signal and the air blowing signal come from different directions, that is, other users use the recording for the audio signal, and at the same time, the other voice signal is used to emit the blowing signal, resulting in the audio signal and The blowing signal is not from the same voice signal, and the audio signal is inconsistent with the direction of the blowing signal. For example, the word "sesame opening" is given, but the pronunciation is not pronounced, so that the direction and blowing of the recording are played. The direction of the gas is inconsistent, and the terminal determines whether the pointing direction feature of the signal portion and the pointing direction characteristic of the audio signal portion are within a preset range to determine whether the audio signal and the blowing signal are from the same directivity direction, thereby avoiding recording attacks. .

For example, the terminal detects that the voice signal of the user A can unlock the terminal, that is, the voice signal of the user A can unlock the terminal. If user B holds the voice recording of user A while blowing the mouth, but does not pronounce it, the direction of playing the recording and the mouth-type blowing may not be the same at this time, but when the ordinary voiceprint is used to unlock the user, user B also The terminal may be successfully unlocked, which poses a security risk and is not safe. In the embodiment of the present invention, it is determined whether the two directivity directions are within a preset range of the microphone array. If the two directivity directions are within a preset range of the microphone array, the sound signals are from the same directivity direction, and There is a recording attack; if the two directivity directions are not within the preset range of the microphone array, the sound signal comes from different directivity directions, and there is a recording attack.

Optionally, determining whether the pointing direction feature of the determining signal portion and the pointing direction feature of the audio signal portion are within a preset range, comprising: respectively determining an angle of a pointing direction of the determining signal portion and a pointing direction of the audio signal portion with a pre-preparation Set the pointing angle threshold comparison.

Optionally, determining the pointing direction feature of the signal portion and the pointing direction feature of the audio signal portion are within a preset range, including: determining that the angle of the pointing direction of the signal portion and the pointing direction of the audio signal portion are smaller than the preset pointing angle Threshold.

In the embodiment of the present invention, the microphone directivity receiving technology can be used to prevent the recording attack. When the terminal recognizes the sound signal, the terminal can include a mode in which the microphone is directed to receive the signal, that is, the microphone enters the directivity listening mode, and only the receiving conforms to the preset. Audio and blow signals in the range of angles avoid recording attacks by limiting the range of microphones that receive audio and blow signals.

The directional reception of the microphone is realized by the sound source localization technology, which can be realized by the microphone array. Generally, the sound transmitted from different directions can be captured, and the microphone is pointed to a specific direction by an algorithm operation to form a “beam” pointing to the sound, and the direction is captured. The audio signal can be used to realize the directionality of the microphone to receive the voice signal. The acoustic waves reach the microhour difference between each microphone in the array, and the microphone array can achieve better directivity than a single microphone. The specific implementation includes that the microphone array can point the sound beam to a range of angles, for example, by a generalized cross-correlation method, smooth coherent transform, phase transform or maximum likelihood, and then adjust the radio direction according to the delay and the set position of the microphone array. Further adjust the microphone finger by adjusting the parameters in the algorithm The tangential receiving direction is a conical shape with an angle θ, and it is further determined that the audio signal and the blowing signal of the received sound source S are both from the effective information number in the direction of the θ1 angle in the direction smaller than the θ angular direction cone.

计算得到声源方向，其中τ为声音到达两个麦克风的延迟量，d为两个麦克风之间的距离，θ1是语音信号的指向性方向角度，C是声音的速度。For example, FIG. 6 is a schematic diagram of an angle of a pointing direction of a sound signal according to Embodiment 1 of the present invention. As shown in FIG. 6, one mobile phone has two microphones A and B, and the distance between A and B is fixed, which is a known d, and the sound propagation speed is fixed to C, and the time difference between the arrival of the microphones A and B according to the sound is τ. The angle θ1 between the sound source (sound signal signal) and the microphone B can be calculated, and it is judged whether it is within the conic shape of the effective sound source direction θ angle according to the angle θ1. Therefore, it can be judged that the audio signal and the air blowing signal of the sound source are effective signals received by the directional microphone. Can also pass the formula

The direction of the sound source is calculated, where τ is the delay amount of the sound reaching the two microphones, d is the distance between the two microphones, θ1 is the directivity direction angle of the speech signal, and C is the speed of the sound.

Further, the distance between the sound source and the microphone can be set. For example, the distance between the sound source and the microphone is determined by the light sensor, the infrared sensor, the ultrasonic sensor, etc., and the distance threshold is set to ensure whether the direction of the recording attack and the blowing signal are consistent. Because if the sound source is close to the microphone, the recording attack and the blow signal will come from the same direction.

S506: determining a sound when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and determining that the pointing direction feature of the signal portion and the pointing direction feature of the audio signal portion are within a preset range The result of the detection is successful.

Specifically, when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and whether the audio signal and the air blowing signal in the voiceprint recognition signal are from the same directivity direction, determining the voiceprint The test result is that the test is successful.

The voiceprint detection method provided by the embodiment of the present invention detects whether there is a sound signal through the terminal. If the terminal detects a sound signal, the terminal receives the sound signal, and the terminal extracts the audio signal portion and the judgment signal portion of the sound signal, and the audio signal portion The voiceprint feature is compared with the preset voiceprint feature to determine whether the pointing direction feature of the signal portion and the pointing direction feature of the audio signal portion are within a preset range, when the voiceprint feature of the audio signal portion and the preset voiceprint feature When the matching degree exceeds the preset threshold, and the pointing direction characteristic of the determined signal portion and the pointing direction characteristic of the audio signal portion are within the preset range, determining that the voiceprint detection result is successful for detection, so that when the terminal recognizes the sound signal, The sound signal is divided into an audio signal part and a judgment signal part to realize double recognition of the sound signal, and at the same time, the situation that the playing recording and the mouth-type blowing direction may be inconsistent is effectively avoided, and the security of the voiceprint unlocking is improved.

FIG. 7 is a flowchart of a method for voiceprint detection according to Embodiment 3 of the present invention. A further implementation manner of the method provided by the embodiment of the present invention, as shown in FIG. 7 , the method provided by the embodiment of the present invention includes:

S701: The terminal detects whether there is a sound signal.

S702: If the terminal detects that there is a sound signal, the terminal receives the sound signal.

S703: The terminal extracts an audio signal portion and a judgment signal portion of the sound signal.

S704: Align the voiceprint feature of the audio signal portion with the preset voiceprint feature.

It should be noted that S701, S702, S703, and S704 are implemented in the same manner as S201, S202, S203, and S204, respectively. For details, refer to the descriptions of S201, S202, S203, and S204, and details are not described herein again.

S705: Compare the perceived temperature characteristic of the determination signal portion with a preset temperature threshold.

Specifically, the determining signal portion may include a sensing temperature characteristic, wherein the sensing temperature characteristic is a temperature when the user corresponding to the sound signal outputs the sound. The sensing temperature characteristic of the determining signal portion is compared with a preset temperature threshold to determine whether the sensing temperature characteristic of the determining signal portion is greater than or equal to a preset temperature threshold. For example, the terminal can sense the temperature of the adjacent microphone through the infrared sensor to determine that the voice signal is from the human body, such as a user, not a recorded electronic device. The preset temperature threshold may be determined according to a temperature range of the human body, and the preset temperature threshold is generally set to a minimum temperature within a normal range of the human body, such as 36 degrees Celsius.

S706: When the matching degree of the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the perceived temperature characteristic of the signal portion is greater than or equal to the preset temperature threshold, determining that the voiceprint detection result is successful.

Specifically, when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the sensing temperature characteristic of the signal portion is greater than or equal to the preset temperature threshold, the voice signal received by the terminal may be determined. Electronic devices from users, not recordings, thus avoiding recording attacks and improving the security of voiceprint unlocking.

The voiceprint detection method provided by the embodiment of the present invention detects whether there is a sound signal through the terminal. If the terminal detects a sound signal, the terminal receives the sound signal, and the terminal extracts the audio signal portion and the judgment signal portion of the sound signal, and the audio signal portion The voiceprint feature is compared with the preset voiceprint feature, and the perceived temperature characteristic of the determination signal portion is compared with the preset temperature threshold, and the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds the preset. Threshold and judge the sense of the signal When the temperature characteristic is greater than or equal to the preset temperature threshold, it is determined that the voiceprint detection result is successful. When the voiceprint feature of the audio signal portion matches the preset voiceprint feature, it is determined whether the perceived temperature characteristic of the signal portion is greater than or equal to The preset temperature threshold is used to determine that the voice signal received by the terminal is from the user, not the recorded electronic device, thereby avoiding the recording attack and improving the security of the voiceprint unlocking.

FIG. 8 is a flowchart of a method for voiceprint detection according to Embodiment 4 of the present invention. The method provided by the embodiment of the present invention is another specific implementation manner of the method provided in the embodiment 1 of FIG. 2 . As shown in FIG. 8 , the method provided by the embodiment of the present invention includes:

S801: When the voiceprint recognition signal is detected, the terminal enters the living voiceprint recognition mode.

Before the voiceprint recognition signal is detected, before the terminal enters the living voiceprint recognition mode, the terminal further includes:

The terminal detects whether there is a voiceprint recognition signal; wherein the voiceprint recognition signal is a sound signal detected when the terminal is in an unlocked state.

The terminal detects whether there is a voiceprint identification signal, including: when the unlocked state, the terminal detects whether there is a sound signal; if the terminal detects a sound signal, the sound signal is a voiceprint recognition signal.

S802: The terminal receives the voiceprint recognition signal and stores it.

S803: The terminal extracts an audio signal portion and a judgment signal portion of the voiceprint recognition signal.

S804: The terminal determines whether the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold. If the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, then S805 is performed; otherwise, S808 is performed.

Optionally, in the implementation of the present invention, the terminal may compare the voiceprint feature of the audio signal portion with the preset voiceprint feature to determine whether the matching relationship between the voiceprint feature of the audio signal portion and the preset voiceprint feature is Exceeded the preset threshold.

S805: The terminal determines whether the audio signal of the audio signal portion and the air blowing signal of the determination signal portion are from the same directivity direction. If the audio signal of the audio signal portion and the air blowing signal of the determination signal portion are from the same directivity direction, then S806 is performed; otherwise, S808 is performed.

Optionally, in the implementation of the present invention, the terminal may determine whether the pointing direction feature of the signal portion and the pointing direction feature of the audio signal portion are within a preset range to determine the audio signal and the judgment signal portion of the audio signal portion. Whether the gas signal comes from the same directivity direction.

S806: The terminal determines whether the text corresponding to the audio signal portion matches the expiratory airflow in the determination signal portion. If the text corresponding to the audio signal portion matches the expiratory airflow in the judgment signal portion, Then, S807 is executed; otherwise, S808 is executed.

Optionally, in the implementation of the present invention, the terminal may compare the expiratory airflow characteristic of the determination signal part with the expiratory airflow characteristic of the audio signal part to determine the text corresponding to the audio signal part and the call in the judgment signal part. Whether the gas flow matches.

S807: The live voiceprint detection was successful.

S808: The live voiceprint detection failed.

It should be noted that, in the implementation of the present invention, after determining that the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, the audio signal and the determination signal of the audio signal portion are determined. Before the partial blowing signal is from the same directivity direction, the method further comprises: determining whether the sensing temperature characteristic of the determining signal portion is greater than or equal to a preset temperature threshold; if the matching relationship between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and determining that the matching degree of the expiratory airflow characteristic of the signal part and the expiratory airflow characteristic of the audio characteristic part exceeds a preset threshold value, and determining a pointing direction characteristic of the signal part and a pointing direction characteristic of the audio signal part are preset The living voiceprint detection is successful within the range and when the perceived temperature characteristic of the signal portion is greater than or equal to the preset temperature threshold.

The voiceprint detection method provided by the embodiment of the present invention, when the voiceprint recognition signal is detected, the terminal enters the living voiceprint recognition mode, the terminal receives the voiceprint recognition signal and stores, and the terminal extracts the audio signal portion of the voiceprint recognition signal. And determining the signal portion, when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the judgment feature of the determination signal portion satisfies a preset determination condition, determining the voiceprint detection result is detection Successfully, when the terminal recognizes the voiceprint recognition signal, the voiceprint recognition signal is divided into an audio signal portion and a judgment signal portion, thereby realizing double recognition of the voiceprint recognition signal, and improving the security of the voiceprint unlocking. At the same time, by matching the voiceprint feature of the audio signal portion with the preset voiceprint feature, and the audio signal in the voiceprint recognition signal and the blow signal are from the same directivity direction, and the audio signal in the voiceprint recognition signal corresponds to When the text matches the exhalation airflow of the blowing signal, it is judged that the voiceprint detection result is successful, and the situation that the playing recording and the mouth-type blowing direction may be inconsistent is effectively avoided, and the security of the voiceprint unlocking is improved.

Further, in the above embodiment, before the terminal extracts the audio signal portion and the determination signal portion of the voiceprint recognition signal, the method further includes:

The terminal separates the voiceprint recognition signal into an audio signal portion and a judgment signal portion;

The terminal separates the voiceprint recognition signal into an audio signal portion and a judgment signal portion, including:

The terminal filters the voiceprint recognition signal by using a filter of a first preset frequency to obtain an audio signal portion;

The terminal filters the voiceprint recognition signal by using a filter of a second preset frequency to obtain a judgment signal part;

The filter of the first preset frequency is a high pass filter, and the filter of the second preset frequency is a low pass filter.

FIG. 9 is a schematic structural diagram of a terminal according to Embodiment 1 of the present invention. As shown in FIG. 9, the terminal provided by the embodiment of the present invention includes: a detecting module 901, a receiving module 902, an extracting module 903, a first matching module 904, and a determining module 905.

The detecting module 901 is configured to detect whether there is a sound signal.

The receiving module 902 is configured to receive a sound signal.

The extraction module 903 is configured to extract an audio signal portion and a determination signal portion of the sound signal.

The first matching module 904 is configured to compare the voiceprint feature of the audio signal portion with the preset voiceprint feature; and compare the expiratory airflow characteristic of the determination signal portion with the expiratory airflow characteristic of the audio signal portion.

The exhalation airflow is characterized by the airflow exhaled when the user corresponding to the sound signal outputs the sound.

The determining module 905 is configured to: when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and determine the matching degree between the expiratory airflow feature of the signal portion and the expiratory airflow feature of the audio feature portion When the preset threshold is exceeded, it is determined that the voiceprint detection result is successful.

The terminal of the embodiment of the present invention is used to perform the technical solution of the method embodiment shown in FIG. 2, and the implementation principle and the technical effect are similar, and details are not described herein again.

Further, in the embodiment shown in FIG. 9, the receiving module 902 is further configured to receive an expiratory airflow feature that is greater than a preset airflow threshold in the determination signal portion.

The terminal further includes: a quantization module.

A quantification module for quantifying the characteristics of the expiratory flow.

The first matching module 904 is further configured to compare the quantized expiratory airflow characteristics with the expiratory airflow characteristics of the audio signal portion.

The matching degree between the expiratory airflow characteristic of the judging signal portion judged by the judging module 905 and the expiratory airflow characteristic of the audio characteristic portion exceeds a preset threshold, including: the quantized expiratory airflow characteristic and the audio The expiratory airflow feature matching of the signal portion exceeds a preset threshold.

Further, in the example shown in FIG. 2, the first matching module 904 is specifically configured to: compare the characteristics of the expiratory airflow with the preset airflow threshold, and if the expiratory airflow characteristic is greater than the preset airflow threshold, The expiratory flow feature is quantized to one; otherwise, the expiratory flow characteristic is quantized to zero.

The matching degree of the expired expiratory airflow characteristic determined by the determining module 905 and the expiratory airflow characteristic of the audio signal portion exceeds a preset threshold, and includes: at least one of the following two situations:

In one case, the expiratory airflow characteristic is quantized to 1, and the text corresponding to the audio signal portion is an aspirated sound.

In another case, the expiratory airflow characteristic is quantized to 0, and the text corresponding to the audio signal portion is an unspised sound.

FIG. 10 is a schematic structural diagram of a terminal according to Embodiment 2 of the present invention. As shown in FIG. 10, the terminal provided by the embodiment of the present invention further includes: a second matching module 906, based on the foregoing embodiment.

The second matching module 906 is configured to determine whether the pointing direction feature of the determining signal portion and the pointing direction feature of the audio signal portion are within a preset range.

The determining mode 905 is further configured to: when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and determine the matching of the expiratory airflow characteristic of the signal portion with the expiratory airflow characteristic of the audio feature portion When the degree exceeds the preset threshold, and the pointing direction characteristic of the signal portion and the pointing direction characteristic of the audio signal portion are within the preset range, it is determined that the voiceprint detection result is the detection success.

The terminal of the embodiment of the present invention is used to implement the technical solution of the method embodiment shown in FIG. 5, and the implementation principle and the technical effect are similar, and details are not described herein again.

Further, in the embodiment shown in FIG. 10, the second matching module 906 is specifically configured to: respectively compare an angle of a pointing direction of the determining signal portion and an angle of a pointing direction of the audio signal portion with a preset pointing angle threshold.

The pointing direction feature of the determining signal portion determined by the determining module 905 and the pointing direction feature of the audio signal portion are within a preset range, including: determining that the angle of the pointing direction of the signal portion and the pointing direction of the audio signal portion are smaller than the preset pointing direction. Angle threshold.

FIG. 11 is a schematic structural diagram of a terminal according to Embodiment 3 of the present invention. As shown in FIG. 11, the terminal provided by the embodiment of the present invention further includes: a third matching module 907, based on the foregoing embodiment.

The third matching module 907 is configured to compare the perceived temperature feature of the determination signal portion with a preset temperature threshold.

The determining module 905 is further configured to: when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and determine the matching of the expiratory airflow characteristic of the signal portion with the expiratory airflow characteristic of the audio feature portion If the degree exceeds the preset threshold, the direction of the pointing direction of the signal portion and the pointing direction characteristic of the audio signal portion are within a preset range, and the sensing temperature characteristic of the signal portion is greater than or equal to the preset temperature threshold, determining the voiceprint detection result is The test was successful.

The terminal of the embodiment of the present invention is used to implement the technical solution of the method embodiment shown in FIG. 7. The implementation principle and the technical effect are similar, and details are not described herein again.

Further, in the embodiment shown in FIG. 11, the terminal further includes: a separation module.

And a separation module, configured to separate the sound signal into an audio signal portion and a determination signal portion before the extraction module extracts the audio signal portion of the sound signal and the determination signal portion.

The separation module is specifically configured to: filter the sound signal by using a filter of a first preset frequency to obtain an audio signal portion; and filter the sound signal by using a filter of a second preset frequency to obtain a judgment signal portion.

The filter of the first preset frequency is a high pass filter, and the filter of the second preset frequency is a low pass filter.

FIG. 12 is a schematic structural diagram of a terminal according to Embodiment 4 of the present invention. As shown in FIG. 12, a terminal provided by an embodiment of the present invention includes: a microphone 1201, a memory 1202, and a processor 1203.

It should be noted that the microphone 1201 may be corresponding to the detection module 901 of the terminal for detecting whether there is a sound signal; if the sound signal is detected, the sound signal is received. The microphone 1503 can also be configured to receive an expiratory airflow characteristic of the determination signal portion that is greater than a preset airflow threshold. The memory 1202 is configured to store execution instructions, and the processor 1203 may be a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or one or more implementations of the embodiments of the present invention. Integrated circuits. When the terminal is running, the processor 1203 communicates with the memory 1202, and the processor 1203 invokes an execution instruction for performing the following operations:

Extracting an audio signal portion and a judgment signal portion of the sound signal; comparing the voiceprint feature of the audio signal portion with the preset voiceprint feature; comparing the expiratory airflow characteristic of the determination signal portion with the expiratory airflow characteristic of the audio signal portion Wherein, the expiratory airflow characteristic is a feature of the airflow exhaled when the user corresponding to the sound signal outputs the sound; when the matching degree of the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the determination signal portion Expiratory airflow characteristics and audio features When the matching degree of the expiratory airflow characteristic exceeds a preset threshold, it is determined that the voiceprint detection result is the detection success.

Optionally, the terminal may further include: a recorder 1204.

It should be noted that the recorder 1204 can be used to collect sound signals emitted by the user, perform feature analysis on the sound signals, and acquire preset voiceprint features and store them.

The processor 1203 is further configured to perform the following operations:

Quantifying the characteristics of the expiratory flow; comparing the quantified expiratory flow characteristics with the characteristics of the expiratory flow of the audio signal portion;

The matching degree between the expiratory airflow characteristic of the judging signal portion determined by the processor 1203 and the expiratory airflow characteristic of the audio characteristic portion exceeds a preset threshold, including: the quantified expiratory airflow characteristic matches the expiratory airflow characteristic of the audio signal portion Degree exceeds the preset threshold

The processor 1203 is further configured to perform the following operations:

Comparing the characteristics of the expiratory airflow with the preset airflow threshold, and if the expiratory airflow characteristic is greater than the preset airflow threshold, quantizing the expiratory airflow characteristic to 1; otherwise, quantifying the expiratory flow characteristic to 0;

The degree of matching between the quantized expiratory airflow characteristic determined by the processor 1203 and the expiratory airflow characteristic of the audio signal portion exceeds a preset threshold, including: at least one of the following two cases:

A case where the expiratory airflow characteristic is quantized to be 1, and the text corresponding to the audio signal portion is an aspirating tone;

In another case, the expiratory airflow characteristic is quantized to 0, and the text corresponding to the audio signal portion is an unspised sound.

The processor 1203 is further configured to perform the following operations:

Determining whether the pointing direction characteristic of the judgment signal portion and the pointing direction characteristic of the audio signal portion are within a preset range; when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and determining the signal portion Judging the voiceprint detection when the matching degree of the expiratory airflow characteristic with the exhalation airflow characteristic of the audio characteristic portion exceeds a preset threshold value, and the pointing direction characteristic of the signal portion and the pointing direction characteristic of the audio signal portion are within a preset range The result is a successful test.

The processor 1203 is further configured to perform the following operations:

Comparing the angle of the pointing direction of the signal portion and the pointing direction of the audio signal portion with the preset pointing angle threshold respectively;

The pointing direction feature of the judgment signal portion judged by the processor 1203 and the pointing direction feature of the audio signal portion are within a preset range, including: determining the angle of the pointing direction of the signal portion and the audio signal The angle of the pointing direction of the number portion is smaller than the preset pointing angle threshold.

The processor 1203 is further configured to perform the following operations:

Comparing the perceived temperature characteristic of the determination signal portion with the preset temperature threshold; when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and determining the expiratory airflow characteristic and audio of the signal portion The matching degree of the expiratory airflow characteristic of the characteristic part exceeds a preset threshold, determining that the pointing direction characteristic of the signal part and the pointing direction characteristic of the audio signal part are within a preset range, and determining that the sensing temperature characteristic of the signal part is greater than or equal to the preset temperature At the threshold, it is judged that the voiceprint detection result is successful.

The processor 1203 is further configured to perform the following operations:

Comparing the voiceprint waveform of the audio signal portion with the preset voiceprint sample feature waveform;

and / or,

The signal frequency of the audio signal portion is compared with the preset voiceprint sample characteristic frequency.

The processor 1203 is further configured to perform the following operations:

The sound signal is separated into an audio signal portion and a judgment signal portion.

Specifically, the sound signal is filtered by using a filter of a first preset frequency to obtain an audio signal portion; and the sound signal is filtered by a filter of a second preset frequency to obtain a judgment signal portion; wherein, the first preset frequency The filter is a high pass filter and the filter of the second preset frequency is a low pass filter.

FIG. 13 is a schematic structural diagram of an apparatus for detecting a voiceprint according to Embodiment 1 of the present invention. The device provided by the example of the present invention can be implemented as a single device, or can be integrated into various voice assistant devices, such as a set top box, a mobile phone, a tablet personal computer, and a laptop computer. , multimedia player, digital camera, personal digital assistant (PDA), navigation device, mobile Internet device (MID) or wearable device (Wearable Device). As shown in FIG. 13, the apparatus provided by the embodiment of the present invention may include one or more of the following units: an input unit, a storage unit, a processor unit, a communication unit, a peripheral interface, an output unit, and a power source.

In the example of the present invention, the microphone can be used as an input unit, and the input unit can input an audio signal to detect whether the terminal has a voiceprint recognition signal. The memory can be used as a storage unit, and the storage unit can store execution instructions, such as an execution instruction such as an operation program and an application program, or a specific blow signal recognition module, a blow signal and an audio signal separation module, and a blow signal determination module. Wait Line instructions. The processor may be a processor unit, and the processor unit may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more implementations of the embodiments of the present invention. Integrated circuits. When the terminal is running, the processor unit communicates with the memory unit, and the processor unit invokes an execution instruction for performing the operations in the above method embodiments. The communication unit can be used for limited or wireless communication between the terminal and other devices. The peripheral interface can be used to provide an interface between the terminal and the peripheral interface module, wherein the peripheral interface module can be a keyboard, a button, or the like. The output unit can be used to output an audio signal. The power supply can be used to provide power to the various units of the terminal.

Embodiments of the present invention also provide a non-transitory computer readable storage medium, such as a storage unit including instructions that are executable by a processor of a voiceprint detecting device to perform the above method. For example, the non-transitory computer readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device.

A non-transitory computer readable storage medium storing computer instructions for causing an apparatus for controlling a cache to perform an operation in the above-described method embodiments. When the instructions in the storage medium are executed by the processor of the terminal, the terminal is enabled to perform the operations in the above method embodiments.

Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, and are not intended to be limiting; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will understand that The technical solutions described in the foregoing embodiments may be modified, or some or all of the technical features may be equivalently replaced; and the modifications or substitutions do not deviate from the technical solutions of the embodiments of the present invention. range.

Claims (22)

A method for detecting voiceprints, comprising: The terminal detects whether there is a sound signal; If the terminal detects the sound signal, the terminal receives the sound signal; The terminal extracts an audio signal portion and a determination signal portion of the sound signal; Comparing the voiceprint feature of the audio signal portion with the preset voiceprint feature; Comparing the expiratory airflow characteristic of the determination signal portion with the expiratory airflow characteristic of the audio signal portion; The exhalation airflow characteristic is a feature of the airflow exhaled when the user corresponding to the sound signal outputs the sound; When the matching degree of the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the matching degree of the expiratory airflow characteristic of the determination signal portion and the expiratory airflow characteristic of the audio feature portion When the preset threshold is exceeded, the voiceprint detection result is determined to be successful. The method of claim 1 further comprising: Receiving, in the portion of the determination signal, the expiratory airflow characteristic that is greater than a preset airflow threshold; Quantifying the characteristics of the expiratory flow; Comparing the quantified expiratory airflow characteristics with the expiratory airflow characteristics of the audio signal portion; The matching degree between the expiratory airflow characteristic of the judging signal portion and the expiratory airflow characteristic of the audio characteristic portion exceeds a preset threshold, and includes: The quantified expiratory airflow characteristic matches the expiratory airflow characteristic of the audio signal portion exceeding a preset threshold. The method of claim 2 wherein said quantifying said expiratory airflow characteristic comprises: Comparing the expiratory airflow characteristic with a preset airflow threshold, and if the exhalation airflow characteristic is greater than the preset airflow threshold, quantizing the exhalation feature to 1; otherwise, The exhalation flow characteristic is quantized to 0; The matching of the quantified expiratory airflow characteristic with the expiratory airflow characteristic of the audio signal portion exceeds a preset threshold, including: at least one of the following two situations: a case where the exhalation airflow characteristic is quantized to 1, and the audio signal portion corresponds to a text This is the air supply tone; In another case, the exhalation airflow feature is quantized to be 0, and the text corresponding to the audio signal portion is a non-aspirated tone. The method according to any one of claims 1 to 3, wherein the method further comprises: Determining whether the pointing direction feature of the determining signal portion and the pointing direction feature of the audio signal portion are within a preset range; And when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the expiratory airflow characteristic of the determination signal portion and the expiratory airflow characteristic of the audio feature portion When the matching degree exceeds the preset threshold, determining that the voiceprint detection result is successful, including: When the matching degree of the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the matching degree of the expiratory airflow characteristic of the determination signal portion and the expiratory airflow characteristic of the audio feature portion When the preset threshold is exceeded, and the pointing direction feature of the determination signal portion and the pointing direction feature of the audio signal portion are within a preset range, the voiceprint detection result is determined to be a successful detection. The method according to claim 4, wherein determining whether the pointing direction feature of the determining signal portion and the pointing direction feature of the audio signal portion are within a preset range comprises: Comparing an angle of the pointing direction of the determining signal portion and an angle of the pointing direction of the audio signal portion with a preset pointing angle threshold, respectively; The pointing direction feature of the determining signal portion and the pointing direction feature of the audio signal portion are within a preset range, including: an angle of a pointing direction of the determining signal portion and an angle of a pointing direction of the audio signal portion are both smaller than The preset points to an angle threshold. The method of claim 5, wherein the method further comprises: Comparing the perceived temperature characteristic of the determination signal portion with a preset temperature threshold; And when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the expiratory airflow characteristic of the determination signal portion and the expiratory airflow characteristic of the audio feature portion When the matching degree exceeds the preset threshold, determining that the voiceprint detection result is successful, including: When the matching degree of the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the expiratory airflow characteristic of the determination signal portion and the expiratory airflow of the audio feature portion The matching degree of the sign exceeds a preset threshold, the pointing direction feature of the determining signal portion and the pointing direction feature of the audio signal portion are within a preset range, and the sensing temperature characteristic of the determining signal portion is greater than or equal to a preset temperature When the threshold is reached, it is determined that the voiceprint detection result is that the detection is successful. The method according to any one of claims 1 to 6, wherein before the terminal extracts the audio signal portion and the determination signal portion of the sound signal, the method further includes: The terminal separates the sound signal into the audio signal portion and the determination signal portion; Separating the voice recognition signal into the audio signal portion and the determination signal portion, the terminal includes: The terminal filters the voice recognition signal by using a filter of a first preset frequency to obtain the audio signal portion; The terminal filters the voice recognition signal by using a filter of a second preset frequency to obtain the judgment signal portion; The filter of the first preset frequency is a high pass filter, and the filter of the second preset frequency is a low pass filter. A terminal, comprising: a detecting module for detecting whether there is a sound signal; a receiving module, configured to receive the sound signal; An extraction module, configured to extract an audio signal portion and a determination signal portion of the sound signal; a first matching module, configured to compare a voiceprint feature of the audio signal portion with a preset voiceprint feature; and perform an expiratory airflow feature of the determination signal portion and an expiratory airflow feature of the audio signal portion Comparison; The exhalation airflow characteristic is a feature of the airflow exhaled when the user corresponding to the sound signal outputs the sound; a judging module, configured to: when the matching degree of the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the expiratory airflow characteristic of the judging signal portion and the exhalation of the audio feature portion When the matching degree of the airflow feature exceeds a preset threshold, the voiceprint detection result is determined to be a successful detection. The terminal of claim 8 wherein: The receiving module is further configured to receive a call that is greater than a preset airflow threshold in the determining signal portion Gas flow characteristics; The terminal further includes: a quantization module; Quantizing module for quantifying the characteristics of the expiratory airflow; The first matching module is further configured to compare the quantized expiratory airflow characteristic with the expiratory airflow characteristic of the audio signal portion; The matching degree of the expiratory airflow characteristic of the judging signal portion determined by the judging module and the expiratory airflow characteristic of the audio characteristic portion exceeds a preset threshold, including: the quantized expiratory airflow characteristic and the The expiratory airflow feature matching degree of the audio signal portion exceeds a preset threshold. The terminal according to claim 9, wherein the first matching module is specifically configured to: Comparing the expiratory airflow characteristic with a preset airflow threshold, and if the exhalation airflow characteristic is greater than the preset airflow threshold, quantizing the exhalation feature to 1; otherwise, The exhalation flow characteristic is quantized to 0; The matching degree between the quantized expiratory airflow characteristic determined by the determining module and the expiratory airflow characteristic of the audio signal portion exceeds a preset threshold, and includes: at least one of the following two situations: a case: the exhalation airflow feature is quantized to be 1, and the text corresponding to the audio signal portion is an aspirating tone; In another case, the exhalation airflow feature is quantized to be 0, and the text corresponding to the audio signal portion is a non-aspirated tone. The terminal according to any one of claims 8 to 10, wherein the terminal further comprises: a second matching module; The second matching module is configured to determine whether a pointing direction feature of the determining signal portion and a pointing direction feature of the audio signal portion are within a preset range; The determining module is further configured to: when the matching degree of the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the expiratory airflow feature of the determining signal portion and the audio feature portion Determining the voiceprint detection result as detection when the matching degree of the expiratory airflow characteristic exceeds a preset threshold, and the pointing direction feature of the determination signal portion and the pointing direction characteristic of the audio signal portion are within a preset range success. The terminal according to claim 11, wherein the second matching module is specifically configured to: Comparing an angle of the pointing direction of the determining signal portion and an angle of the pointing direction of the audio signal portion with a preset pointing angle threshold, respectively; The pointing direction feature of the determining signal portion and the pointing direction feature of the audio signal portion determined by the determining module are within a preset range, including: an angle of a pointing direction of the determining signal portion and a portion of the audio signal portion The angles of the pointing directions are all smaller than the preset pointing angle threshold. The terminal according to claim 12, wherein the terminal further comprises: a third matching module; a third matching module, configured to compare the sensing temperature feature of the determining signal portion with a preset temperature threshold; The determining module is further configured to: when the matching degree of the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the expiratory airflow feature of the determining signal portion and the audio feature portion The matching degree of the expiratory airflow characteristic exceeds a preset threshold, the pointing direction characteristic of the judging signal portion and the pointing direction characteristic of the audio signal portion are within a preset range, and the sensing temperature characteristic of the judging signal portion is greater than When the preset temperature threshold is equal to the preset temperature threshold, it is determined that the voiceprint detection result is that the detection is successful. The terminal according to any one of claims 8 to 13, wherein the terminal further comprises: a separation module; The separating module is configured to separate the sound signal into the audio signal portion and the determining signal portion before the extracting module extracts an audio signal portion and a determining signal portion of the sound signal; The separation module is specifically configured to: And the sound signal is filtered by using a filter of a first preset frequency to obtain the audio signal portion; and the sound signal is filtered by using a filter of a second preset frequency to obtain the determination signal portion; The filter of the first preset frequency is a high pass filter, and the filter of the second preset frequency is a low pass filter. A terminal, comprising: a microphone and a processor; The microphone is configured to detect whether there is a sound signal; if the sound signal is detected, the sound signal is received; The processor is configured to extract an audio signal portion and a judgment signal portion of the sound signal; Comparing the voiceprint feature of the audio signal portion with a preset voiceprint feature; comparing the expiratory airflow characteristic of the determination signal portion with the expiratory airflow characteristic of the audio signal portion; wherein The expiratory airflow characteristic is a feature of the airflow exhaled when the user corresponding to the sound signal outputs the sound; when the matching degree of the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold, and the determining When the matching degree of the expiratory airflow characteristic of the signal part and the expiratory airflow characteristic of the audio characteristic part exceeds a preset threshold, determining that the voiceprint detection result is successful detection. The terminal of claim 15 wherein: The microphone is further configured to receive the expiratory airflow characteristic of the determination signal portion that is greater than a preset airflow threshold; The processor is further configured to quantize the characteristics of the expiratory airflow; and compare the quantized expiratory airflow characteristics with the expiratory airflow characteristics of the audio signal portion; The matching degree of the expiratory airflow characteristic of the judging signal portion and the expiratory airflow characteristic of the audio characteristic portion determined by the processor exceeds a preset threshold, and includes: The quantified expiratory airflow characteristic matches the expiratory airflow characteristic of the audio signal portion exceeding a preset threshold. The terminal according to claim 16, wherein the processor is specifically configured to: Comparing the expiratory airflow characteristic with a preset airflow threshold, and if the exhalation airflow characteristic is greater than the preset airflow threshold, quantizing the exhalation feature to 1; otherwise, The exhalation flow characteristic is quantized to 0; The matching between the quantized expiratory airflow characteristic determined by the processor and the expiratory airflow characteristic of the audio signal portion exceeds a preset threshold, and includes: at least one of the following two situations: a case: the exhalation airflow feature is quantized to be 1, and the text corresponding to the audio signal portion is an aspirating tone; In another case, the exhalation airflow feature is quantized to be 0, and the text corresponding to the audio signal portion is a non-aspirated tone. A terminal according to any one of claims 15-17, characterized in that The processor is further configured to determine whether a pointing direction feature of the determining signal portion and a pointing direction feature of the audio signal portion are within a preset range; when the voice signal portion has a voiceprint feature and a preset voiceprint The matching degree of the feature exceeds a preset threshold, and the matching degree of the expiratory airflow characteristic of the judging signal portion with the expiratory airflow characteristic of the audio characteristic portion exceeds a preset threshold, and When the pointing direction feature of the determining signal portion and the pointing direction feature of the audio signal portion are within a preset range, determining that the voiceprint detecting result is successful for detecting. The terminal according to claim 18, wherein the processor is specifically configured to: Comparing an angle of the pointing direction of the determining signal portion and an angle of the pointing direction of the audio signal portion with a preset pointing angle threshold, respectively; The pointing direction feature of the determining signal portion and the pointing direction feature of the audio signal portion determined by the processor are within a preset range, including: an angle of a pointing direction of the determining signal portion and a portion of the audio signal portion The angles of the pointing directions are all smaller than the preset pointing angle threshold. The terminal according to claim 19, characterized in that The processor is further configured to compare the sensing temperature feature of the determining signal portion with a preset temperature threshold; when the matching degree between the voiceprint feature of the audio signal portion and the preset voiceprint feature exceeds a preset threshold And the matching degree of the expiratory airflow characteristic of the judging signal portion and the expiratory airflow characteristic of the audio characteristic portion exceeds a preset threshold, the pointing direction characteristic of the judging signal portion and the pointing direction of the audio signal portion When the feature is within the preset range, and the sensing temperature characteristic of the determining signal portion is greater than or equal to the preset temperature threshold, determining that the voiceprint detection result is successful. A terminal according to any one of claims 15 to 20, characterized in that The processor is further configured to separate the sound signal into the audio signal portion and the determination signal portion; The processor is specifically configured to: And the sound recognition signal is filtered by using a filter of a first preset frequency to obtain the audio signal portion; and the sound recognition signal is filtered by using a filter of a second preset frequency to obtain the determination signal portion; The filter of the first preset frequency is a high pass filter, and the filter of the second preset frequency is a low pass filter. A non-transitory computer readable storage medium, wherein the non-transitory computer readable storage medium stores computer instructions for causing a device that controls a cache to perform a claim 1-7 Any of the methods described.

Images