[go: up one dir, main page]

CN102541502A - Method and system for outputting pronunciation character - Google Patents

Method and system for outputting pronunciation character Download PDF

Info

Publication number
CN102541502A
CN102541502A CN2010106051935A CN201010605193A CN102541502A CN 102541502 A CN102541502 A CN 102541502A CN 2010106051935 A CN2010106051935 A CN 2010106051935A CN 201010605193 A CN201010605193 A CN 201010605193A CN 102541502 A CN102541502 A CN 102541502A
Authority
CN
China
Prior art keywords
character
sound
height
cell
loudness
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN2010106051935A
Other languages
Chinese (zh)
Other versions
CN102541502B (en
Inventor
马宇尘
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Liangming Technology Development Co Ltd
Original Assignee
Shanghai Liangming Technology Development Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Liangming Technology Development Co Ltd filed Critical Shanghai Liangming Technology Development Co Ltd
Priority to CN201010605193.5A priority Critical patent/CN102541502B/en
Publication of CN102541502A publication Critical patent/CN102541502A/en
Application granted granted Critical
Publication of CN102541502B publication Critical patent/CN102541502B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Stereophonic System (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

The invention provides a method and system for outputting a pronunciation character, and belongs to the technical field of computers and software. The method comprises the following steps of: 1, collecting a character to be processed and sound information corresponding to the character; 2, separating the character to be processed into character units, and distributing the sound information corresponding to the character units; 3, comparing and determining heights of the character units according to a height mode of at least one of pitch and intensity of the sound information; and 4, outputting a character with the height adjusted according to the height mode. According to the method and the system for outputting the pronunciation character, a relationship between description of the pitch, intensity or other elements of sound and the output character is directly established, and convenience is brought to people when people understand or study a character to be pronounced.

Description

The method and system of pronunciation character output
Technical field
The invention belongs to computing machine, software technology field.
Technical background
In daily work and life, we through regular meeting output is various can be in order to the character style of pronunciation, such as; Our said at ordinary times various language; The various songs of being sung, the various literal of being imported, and the various printed matters of being seen; These contents, they all are the character styles that can pronounce.
The describing mode of sound, the describing mode that we often have one type of pitch or a loudness of a sound is described the situation of sound.Described pitch also can be called audio frequency, in order to describe the sound height of frequency of user.Described loudness of a sound mainly refers to the intensity that the user sounds, and is that unit is described usually with the decibel.
We the character style that can see, if can the output state and the factors such as pitch or loudness of a sound of its character be connected, so, will give people understand and study pronunciation character in, offer convenience.
Summary of the invention
The object of the invention provides a kind of method and system of the character output of pronouncing, and utilizes the present invention, can factors such as the pitch of describing sound or loudness of a sound be contacted directly with the character of output.
A kind of method of the character output of pronouncing, this method includes following steps:
Step 1 is gathered pending character, and the acoustic information corresponding with character;
Step 2 becomes character cell with pending Character segmentation, distributes the acoustic information corresponding with character cell;
Step 3, according to both height patterns of one of which at least of the pitch of acoustic information or loudness of a sound, the height of character cell is judged in comparison;
Step 4, output are pressed the character behind the height mode adjustment height.
Method described in the invention also has following technical characterictic:
Further, described pending character is the pairing literal of language.
Further, corresponding described step 2 or step 3, with each independently literal as a character cell.
Further, can be that the basis is done and decomposed a plurality of character cells with the syllable with same word, with each independently syllable as a character cell.
Further, with phrase, the phrase that a plurality of word is formed, the assembly that does not perhaps have the word of relevance to form is as character cell.
Further, corresponding described step 2, the allocation scheme between sound and the character cell includes one of following mode,
(1) utilizes relevance between sound rhythm and the character rhythm, come to be the character cell assigned sound;
(2) carry out speech recognition, with the result after the speech recognition, compared with the form and the existing character cell of identification character, then, according to the result of comparison, the speech recognition character just can be mapped with each character cell;
(3) under corresponding the situation of character cell, import corresponding sound;
(4) adopt the syllable identification mode to distribute.
Further, corresponding described mode (1), it includes following steps,
According to the tempo variation situation of sound aspect, earlier sound is partly cut apart;
Information after will cutting apart is compared with the corresponding characters unit;
Give character cell with supporting acoustic information according to comparison result.
Further, corresponding described mode (1), it includes following steps,
Select the specific syllable of pronunciation information;
Compare with the characteristic syllable of character;
Under the situation that obtains to confirm in comparison, with corresponding between corresponding sound and the character cell.
Further, when judging pitch or loudness of a sound numerical value, for the situation that is comprising different pitches or loudness of a sound numerical value in the same character cell, processing mode comprises following type one of which at least,
(1) gets the mxm. of pitch in this character cell or loudness of a sound, be used as the data of pitch in this character cell or loudness of a sound;
(2) with the mxm. and the minimum of pitch in this character cell or loudness of a sound, carry out data average after, be used as the data of the pitch or the loudness of a sound of this character cell;
(3) with the pairing time period of character cell, be divided into a plurality of independently ingredients, carry out selectivity and handle;
(4), pitch that records or loudness of a sound numerical value are averaged with after the character cell staging treating.
Further, corresponding described mode (3), in the quantity of ingredient during more than or equal to 3, operation steps does,
The numerical value of pitch or loudness of a sound in elder generation's these word time sections of measurement;
Remove the mxm. and the minimum of pitch or loudness of a sound again;
And then get the mean value of the corresponding numerical value of each ingredient of residue.
Further, corresponding described step 3 or step 4, when the numerical value of the pitch of acoustic information or loudness of a sound was big, the height of the corresponding character cell of institute is height just, otherwise, the height of corresponding character cell just low.
Further, corresponding described step 3 or step 4, the character height that is obtained is concrete numerical value, or ratio.
Further, corresponding described step 3 or step 4 with the character cell of the predetermined fraction of required processing, are got its mean value, are used as with reference to height.
Further, corresponding described step 3 or step 4, the way of output of being carried out, for following type one of them,
(1) height in the character bottom is under the situation of unanimity, with the height of font unit, upwards does the mode of extending or compressing;
(2) mode of the line of reference that floats is set, with the centre position of each character, perhaps other preposition is corresponding with this floating neutral;
(3) can adopt the highly mode of type baseline, the height of a character is set.
Further, described mode (3), its implementation process includes following steps,
Steps A 1 is judged the pitch of each character cell or the numerical value of loudness of a sound;
Steps A 2, at the pairing coordinate position of this character cell place, be provided with corresponding height with reference to baseline;
Steps A 3 is got the special ratios value, on this height coordinate, location point is set;
Steps A 4 moves to the location point place that step 3 sets with the pairing character cell of each voice data.
The present invention also provides a kind of system of the character output of pronouncing, and this system includes following ingredient:
Acquisition module is gathered pending character, and the acoustic information corresponding with character;
Cut apart module, pending Character segmentation is become character cell, distribute the acoustic information corresponding with character cell;
Judge module, according to both height patterns of one of which at least of the pitch of acoustic information or loudness of a sound, the height of character cell is judged in comparison;
Output module, output are pressed the character behind the height mode adjustment height.
Description of drawings
Fig. 1 is a pronunciation character described in the invention, is utilizing the present invention to handle and is exporting synoptic diagram before.
Fig. 2 utilizes method of the present invention, corresponding the character that Fig. 1 provided, and the synoptic diagram of the pronunciation character after handling is a kind of embodiment.
Fig. 3 utilizes method of the present invention, corresponding the character that Fig. 1 provided, and the synoptic diagram of the pronunciation character after handling is another kind of embodiment.
Fig. 4-the 1st utilizes method of the present invention, and the structural representation when highly handling is another kind of embodiment.
Fig. 4-the 2nd utilizes method of the present invention, the synoptic diagram after carrying out highly handling, correspondence the described embodiment of Fig. 4-1.
Fig. 5 is the process flow diagram of method described in the invention.
Fig. 6 is the structured flowchart of system according to the invention.
Specific embodiment
For ease in to understanding of the present invention, below as for example and non-limiting, thought of the present invention is described:
Various character styles, when pronouncing, can be corresponding corresponding pitch or loudness of a sound; The expressing information of these sound; Can be corresponding with the output mode of character, so, just can be through the output mode of character; Particularly the height of character output is expressed the pitch or the loudness of a sound information of sound.
Below in conjunction with Fig. 1, Fig. 2, Fig. 3, Fig. 4-1, Fig. 4-2, and shown in Figure 5, method described in the invention is elaborated.
Method described in the invention mainly includes following steps:
Step 1 is gathered pending character, and the acoustic information corresponding with character.
In this step, we need be to those pending character type information, and the acoustic information corresponding with character carry out acquisition operations.Pending character information is made a general reference various character styles.Can be corresponding different language form, except the pairing literal commonly used of language, can also correspondence can realize other character style of voice effect, perhaps have other sign format of character function.
Join shown in Figure 1ly, pending character 100 usefulness square forms are represented.
Corresponding pending character needs to distribute corresponding acoustic information.We need gather these acoustic informations, in order to be converted into the data of corresponding pitch or loudness of a sound aspect.
Step 2 becomes character cell with pending Character segmentation, distributes the acoustic information corresponding with character cell.
The character style of required processing need be divided into the character cell with relative independentability.Each character cell is made as a whole data processing of carrying out pitch or loudness of a sound aspect.
Concerning literal, each word just can be regarded as a character cell.In addition, can also the quantity of same radical according to syllable be resolved into a plurality of character cells, such as, with each independently syllable also be fine as a character cell.
In addition, phrase, the phrase that can also a plurality of word be formed, the assembly that does not perhaps have the word of relevance to form also can be regarded as character cell.
As preferred embodiment, can with each independently word as a character cell independently.
Be directed to each character cell, all corresponding acoustic information can be arranged.We need cut apart and distribute these acoustic informations, understand each character cell the pitch or the loudness of a sound of corresponding acoustic information.
To the allocation scheme between sound and the character cell, non-limiting as giving an example, following several kinds of forms are arranged:
(1) utilizes relevance between sound rhythm and the character rhythm, come to be the character cell assigned sound.
Between the pairing sound of each character cell, the variation of corresponding rhythm aspect can be arranged all.Therefore, can earlier sound partly be cut apart according to the tempo variation situation of sound aspect, then, the information after will cutting apart is compared with the corresponding characters unit, gives character cell with supporting acoustic information according to comparison result again.
(2) carry out speech recognition, with the result after the speech recognition, compared with the form and the existing character cell of identification character, then, according to the result of comparison, the speech recognition character just can be mapped with each character cell.
Then, with the pairing pronunciation information of speech recognition character, corresponding again with each character cell.Utilize this mode, also can acoustic information and corresponding character cell be mapped.
(3) can compare with the mode of artificial selection, the user can be directed to corresponding sound, is contrasting to import corresponding character, so sound and character also can be mapped.
(4) can under the situation of character cell in correspondence, import corresponding sound.So, also can well character and sound be combined.The mode of this combination is the synchronism through sound and literal, carries out effectively distributing between acoustic information and the character cell.
(5) adopt the syllable identification mode to distribute.
Can also select the specific syllable of pronunciation information, compare, need corresponding vocabulary not discerned with the characteristic syllable of character.
Such as, when having comprised the pronunciation key element of " B " in the pronunciation information, just only need the content that comprise this syllable in the character be searched for, then, if can directly confirm, so, just can be with having carried out correspondence between corresponding sound and the character cell;
If can not directly confirm, so, can be used as confirming factor through the pronunciation of other adjacent part, effectively judge.
Step 3, according to both height patterns of one of which at least of the pitch of acoustic information or loudness of a sound, the height of character cell is judged in comparison.
In this step, the acoustic information that we need distribute preceding step, and compare each other between the character cell.Result according to comparison judges the height situation that character cell should be provided with.
The height pattern of acoustic information, refer to different audio the height allocative decision of corresponding character.
In the present invention, we adopt pitch or loudness of a sound both at least the height pattern of one of which judge.Therefore, to the pairing pronunciation information of each character cell, we need judge the numerical value of its pitch or loudness of a sound.
It is pointed out that when judging pitch or loudness of a sound numerical value, possibly comprise different pitches or loudness of a sound numerical value in the same character cell.In this case, we need carry out data processing to measuring pitch or the loudness of a sound numerical value that obtains.
Several kinds of processing schemes are like this arranged:
(1) gets the mxm. of pitch in this character cell or loudness of a sound, be used as the data of pitch in this character cell or loudness of a sound.
(2) with the mxm. and the minimum of pitch in this character cell or loudness of a sound, carry out data average after, be used as the data of the pitch or the loudness of a sound of this character cell.
(3) with the pairing time period of character cell, be divided into a plurality of independently ingredients, the quantity of this ingredient is more than or equal to 3.When carrying out, be such: the numerical value of pitch or loudness of a sound in elder generation's these word time sections of measurement, remove the mxm. and the minimum of pitch or loudness of a sound again, and then get the mean value of the corresponding numerical value of each ingredient of residue.
(4), pitch that records or loudness of a sound numerical value are averaged with after the character cell staging treating.
(5) other treating method.Because the mode that data are handled has many kinds, just no longer give an example here.
Certainly, final processing mode does not limit.
In the present invention, we have a cardinal rule is such: when the numerical value of the pitch of acoustic information or loudness of a sound is big, the height of corresponding character cell just high; Otherwise, the height of corresponding character cell just low.
In this case, we can set a preset pattern, judge the pairing height of each character cell.
Below, and non-limiting, we provide for reference character height determinating mode as for example:
Adopt the different pitches or the numerical range of loudness of a sound, corresponding the mode of respective symbols height is set up judgment model.We only need compare the pitch or the loudness of a sound numerical value of acquisition sound with this pattern, just can directly obtain the corresponding height of character.
The character height that is obtained can be concrete numerical value, also can be a ratio.
Such as with the pitch or the loudness of a sound of some numerical value, it highly is to be 1, the numerical value of remainder sound, if higher than these data, such as, high by 20 percent, so, the height of character also can exceed 20% with respect to this with reference to numerical value; Certainly, if lower by 20 percent than reference pattern, so, the character height that is appeared also can hang down 20%.
Further, can also the character cell of the predetermined fraction of required processing be got its mean value, be used as with reference to height.This predetermined fraction can be in short, also can be one section word, also can be entire article; Specifically do not limit.
When having obtained corresponding height numerical value, need export with the corresponding way of output.
In the present invention, it is for reference just to provide the multiple way of output:
(1) height in the character bottom is under the situation of unanimity, with the height of font unit, upwards does the mode of extending or compressing.Join shown in Figure 2ly, the embodiment that is showed here is exactly this form.It is pointed out that when highly increasing or reduce, the transverse width of corresponding font, also be fit to do corresponding adjustment; Certainly, not making change also is fine.
(2) mode of the line of reference that floats is set, such as, join shown in Figure 3ly, can be provided with the line of reference 200 that floats, in the present embodiment, be that the mode of employing center line realizes.Then, with the centre position of each character, perhaps other preposition is corresponding with this floating neutral, so the pitch of each character or loudness of a sound just can have been checked in this way.
(3) can adopt the highly mode of type baseline, the height of a character is set.
Shown in ginseng Fig. 4-2, set with reference to baseline 300 in this embodiment, be in advance according to the size of pitch or loudness of a sound, the height number that is provided with.Between the correspondence position of each height number, such as, between the summit, can make corresponding line, also can not do line.
Then, the size of each character cell can be by with a kind of font size; Certainly, also can make differentiation.Then, it is distributed on the baseline with reference to effect; The vertex position place of baseline particularly.
Utilize this mode, we also can check the numerical value of pairing pitch of each character cell or loudness of a sound well.
Further, Fig. 4-1 that we combine can also describe a kind of in order to realize the process of this mode.This process includes following steps:
Steps A 1 is judged the pitch of each character cell or the numerical value of loudness of a sound;
Steps A 2, at the pairing coordinate position of this character cell place, be provided with corresponding height with reference to baseline 300;
Steps A 3 is got the special ratios value, on this height coordinate, location point is set;
Steps A 4 moves to the location point place that step 3 sets with the pairing character cell of each voice data.
So, handle through similar mode, just can accomplish the output of corresponding data.
Step 4, output are pressed the character behind the height mode adjustment height.
With the data that previous processed is crossed, just can utilize the mode that is shown like Fig. 2, Fig. 3, Fig. 4-2 to export.Certainly, also can adopt other form.
The present invention can be applied in various occasions of carrying out character output, such as, when carrying out lyrics expression, just can express the sound variation of the lyrics better; Certainly, for language learning, this mode also has extensive value.
Further, utilize aforesaid method, the present invention also provides a kind of system of the character output of pronouncing) this system includes following ingredient:
Acquisition module 400 is gathered pending character, and the acoustic information corresponding with character;
Cut apart module 410, pending Character segmentation is become character cell, distribute the acoustic information corresponding with character cell;
Judge module 420, according to both height patterns of one of which at least of the pitch of acoustic information or loudness of a sound, the height of character cell is judged in comparison;
Output module 430, output are pressed the character behind the height mode adjustment height.
More than be the description of this invention and non-limiting, based on other embodiment of inventive concept, also all among protection scope of the present invention.

Claims (16)

1. the method for the character output of pronouncing is characterized in that this method includes following steps:
Step 1 is gathered pending character, and the acoustic information corresponding with character;
Step 2 becomes character cell with pending Character segmentation, distributes the acoustic information corresponding with character cell;
Step 3, according to both height patterns of one of which at least of the pitch of acoustic information or loudness of a sound, the height of character cell is judged in comparison;
Step 4, output are pressed the character behind the height mode adjustment height.
2. the method for a kind of character output of pronouncing according to claim 1, it is characterized in that: described pending character is the pairing literal of language.
3. the method for a kind of character output of pronouncing according to claim 2 is characterized in that: corresponding described step 2 or step 3, with each independently literal as a character cell.
4. the method for a kind of character output of pronouncing according to claim 1 is characterized in that: can be that the basis is done and decomposed a plurality of character cells with the syllable with same word, with each independently syllable as a character cell.
5. the method for a kind of character output of pronouncing according to claim 1 is characterized in that: with phrase, the phrase that a plurality of word is formed, the assembly that does not perhaps have the word of relevance to form is as character cell.
6. the method for a kind of character output of pronouncing according to claim 1 is characterized in that: corresponding described step 2, and the allocation scheme between sound and the character cell includes one of following mode,
(1) utilizes relevance between sound rhythm and the character rhythm, come to be the character cell assigned sound;
(2) carry out speech recognition, with the result after the speech recognition, compared with the form and the existing character cell of identification character, then, according to the result of comparison, the speech recognition character just can be mapped with each character cell;
(3) under corresponding the situation of character cell, import corresponding sound;
(4) adopt the syllable identification mode to distribute.
7. the method for a kind of character output of pronouncing according to claim 6 is characterized in that: corresponding described mode (1), and it includes following steps,
According to the tempo variation situation of sound aspect, earlier sound is partly cut apart;
Information after will cutting apart is compared with the corresponding characters unit;
Give character cell with supporting acoustic information according to comparison result.
8. the method for a kind of character output of pronouncing according to claim 6 is characterized in that: corresponding described mode (1), and it includes following steps,
Select the specific syllable of pronunciation information;
Compare with the characteristic syllable of character;
Under the situation that obtains to confirm in comparison, with corresponding between corresponding sound and the character cell.
9. the method for a kind of character output of pronouncing according to claim 1; It is characterized in that: when judging pitch or loudness of a sound numerical value; For the situation that is comprising different pitches or loudness of a sound numerical value in the same character cell, processing mode comprises following type one of which at least
(1) gets the mxm. of pitch in this character cell or loudness of a sound, be used as the data of pitch in this character cell or loudness of a sound;
(2) with the mxm. and the minimum of pitch in this character cell or loudness of a sound, carry out data average after, be used as the data of the pitch or the loudness of a sound of this character cell;
(3) with the pairing time period of character cell, be divided into a plurality of independently ingredients, carry out selectivity and handle;
(4), pitch that records or loudness of a sound numerical value are averaged with after the character cell staging treating.
10. the method for a kind of character output of pronouncing according to claim 9 is characterized in that: corresponding described mode (3), and in the quantity of ingredient during more than or equal to 3, operation steps does,
The numerical value of pitch or loudness of a sound in elder generation's these word time sections of measurement;
Remove the mxm. and the minimum of pitch or loudness of a sound again;
And then get the mean value of the corresponding numerical value of each ingredient of residue.
11. the method for a kind of character output of pronouncing according to claim 1; It is characterized in that: corresponding described step 3 or step 4, when the numerical value of the pitch of acoustic information or loudness of a sound was big, the height of the corresponding character cell of institute is height just; Otherwise, the height of corresponding character cell just low.
12. the method for a kind of character output of pronouncing according to claim 1 is characterized in that: corresponding described step 3 or step 4, the character height that is obtained is concrete numerical value, or ratio.
13. the method for a kind of character output of pronouncing according to claim 1 is characterized in that: corresponding described step 3 or step 4, with the character cell of the predetermined fraction of required processing, get its mean value, be used as with reference to height.
14. the method for a kind of character output of pronouncing according to claim 1 is characterized in that: corresponding described step 3 or step 4, the way of output of being carried out, for following type one of them,
(1) height in the character bottom is under the situation of unanimity, with the height of font unit, upwards does the mode of extending or compressing;
(2) mode of the line of reference that floats is set, with the centre position of each character, perhaps other preposition is corresponding with this floating neutral;
(3) can adopt the highly mode of type baseline, the height of a character is set.
15. the method for a kind of character output of pronouncing according to claim 14 is characterized in that: described mode (3), its implementation process includes following steps,
Steps A 1 is judged the pitch of each character cell or the numerical value of loudness of a sound;
Steps A 2, at the pairing coordinate position of this character cell place, be provided with corresponding height with reference to baseline;
Steps A 3 is got the special ratios value, on this height coordinate, location point is set;
Steps A 4 moves to the location point place that step 3 sets with the pairing character cell of each voice data.
16. the system of the character output of pronouncing is characterized in that this system includes following ingredient:
Acquisition module is gathered pending character, and the acoustic information corresponding with character;
Cut apart module, pending Character segmentation is become character cell, distribute the acoustic information corresponding with character cell;
Judge module, according to both height patterns of one of which at least of the pitch of acoustic information or loudness of a sound, the height of character cell is judged in comparison;
Output module, output are pressed the character behind the height mode adjustment height.
CN201010605193.5A 2010-12-25 2010-12-25 The method and system that pronunciation character exports Active CN102541502B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201010605193.5A CN102541502B (en) 2010-12-25 2010-12-25 The method and system that pronunciation character exports

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201010605193.5A CN102541502B (en) 2010-12-25 2010-12-25 The method and system that pronunciation character exports

Publications (2)

Publication Number Publication Date
CN102541502A true CN102541502A (en) 2012-07-04
CN102541502B CN102541502B (en) 2016-02-03

Family

ID=46348481

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201010605193.5A Active CN102541502B (en) 2010-12-25 2010-12-25 The method and system that pronunciation character exports

Country Status (1)

Country Link
CN (1) CN102541502B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111063372A (en) * 2019-12-30 2020-04-24 广州酷狗计算机科技有限公司 Method, device and equipment for determining pitch characteristics and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1570858A (en) * 2003-07-16 2005-01-26 英业达股份有限公司 Editing system for accompaniment subtitles and its editing and displaying method
JP2006268879A (en) * 2006-05-29 2006-10-05 Fujitsu Ltd Data processing apparatus and audio output control method
US20090300277A1 (en) * 2008-05-27 2009-12-03 Micron Technology, Inc. Devices and methods for operating a solid state drive
CN101751914A (en) * 2008-12-04 2010-06-23 江亮都 Lyric display system and method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1570858A (en) * 2003-07-16 2005-01-26 英业达股份有限公司 Editing system for accompaniment subtitles and its editing and displaying method
JP2006268879A (en) * 2006-05-29 2006-10-05 Fujitsu Ltd Data processing apparatus and audio output control method
US20090300277A1 (en) * 2008-05-27 2009-12-03 Micron Technology, Inc. Devices and methods for operating a solid state drive
CN101751914A (en) * 2008-12-04 2010-06-23 江亮都 Lyric display system and method

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111063372A (en) * 2019-12-30 2020-04-24 广州酷狗计算机科技有限公司 Method, device and equipment for determining pitch characteristics and storage medium

Also Published As

Publication number Publication date
CN102541502B (en) 2016-02-03

Similar Documents

Publication Publication Date Title
EP0960376A4 (en) Text processing and retrieval system and method
CN104347080A (en) Voice analysis method and device, voice synthesis method and device, and medium storing voice analysis program
CN1889170B (en) Method and system for generating synthesized speech based on recorded speech template
IL172518A0 (en) System and method for configuring voice readers using semantic analysis
WO2006023622A3 (en) Automated extraction of semantic content and generation of a structured document from speech
CN105161116B (en) The determination method and device of multimedia file climax segment
CN105138515A (en) Named entity recognition method and device
WO2021151344A1 (en) Somethod and apparatus for song synthesis, and computer readable storage medium
WO2002001404A3 (en) Automated generation of text analysis systems
CN106297764A (en) A kind of multilingual mixed Chinese language treatment method and system
CN106297766B (en) Phoneme synthesizing method and system
CN101963972A (en) Method and system for extracting emotional keywords
CN101615417A (en) A kind of Chinese synchronously displaying lyrics method that is accurate to word
CN107220339A (en) A kind of lyrics word for word display methods
US7722082B2 (en) System and method for inserting numeric labels automatically
CN103699598A (en) Method for displaying data reference process in graphical manner and device thereof
CN102541502A (en) Method and system for outputting pronunciation character
Post French tonal structures
CN106503181B (en) Audio data processing method and device
CN102063292B (en) Harmony compiling method and device
CN104766602B (en) Sing fundamental frequency synthetic parameters generation method and system in synthesis system
CN110321404A (en) Entry choosing method, device, electronic equipment and the storage medium of study of words
JP2003233825A (en) Document processor
CN102081677B (en) Generation method of long sentences in Pinyin input method
CN110120216B (en) Audio data processing method and device for singing evaluation

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant