FR3011647A1 - METHOD FOR CONTROLLING EQUIPMENT CONNECTED TO A COMMUNICATION NETWORK, EQUIPMENT, SYSTEM, COMPUTER PROGRAM PRODUCT AND CORRESPONDING DATA CARRIER - Google Patents

Abstract

The invention relates to a method for controlling a first device connected to a communication network and controllable by at least one command. According to the invention, the method comprises searching for an association of at least one piece of information representative of an audio sequence, acquired by at least a second piece of equipment connected to said network, with at least one of said commands, said candidate command, said search being implemented at least partially by said first equipment.

Description

Method of controlling equipment connected to a communication network, equipment, system, computer program product and corresponding data carrier 1. Field of the invention The field of disclosure relates to the control of communicating objects, in particular communicating objects connected to a network, for example a home network. A method of controlling equipment connected to a data network (and a corresponding equipment, system, computer program product and data carrier) is described. In particular, the method and the system may allow the control of a piece of equipment of a home network, for example a data sharing network. 2. Prior art A home environment includes more and more communicating objects, made available to a user in his daily life. For example, they may be objects traditionally dedicated to sharing multimedia content such as tablets, smart phones, televisions, decoders, gateways connecting to a local area network, equipment for acquiring data. physical devices (for example temperature sensors, pressure sensors, presence detectors, etc.) or equipment to be controlled having other functions such as household appliances (such as so-called "smart refrigerators") allowing in particular to offer new services to a user, for example by means of sensors and means of providing information to a user. Some of these objects may offer only limited means to control them because of their small size or low market value or because they are more likely to be providers of information. For example, some objects may not have a graphical interface. Some objects can be controlled through a more complete interface but difficult to apprehend for a user, for example because the commands are complex or not ergonomic. Thus, there is a need to allow a user to have a user-friendly interface when he wants to use a communicating object. SUMMARY OF THE INVENTION The invention makes it possible to improve the situation by proposing a method that is more user-friendly for a user of a device to be controlled than the solutions of the prior art.

More particularly, the invention relates to a method of controlling a first device connected to a communication network and controllable by at least one command. In particular, in certain embodiments, said method comprises: a search for an association of at least one piece of information representative of an audio sequence, acquired by at least one second piece of equipment connected to said network, with at least one of said commands , called candidate command. For example, according to a first particular embodiment, said method comprises: a reception, at the level of the first equipment, of at least one audio sequence acquired by at least a second equipment connected to said network, a search of an association said audio sequence acquired with at least one of said commands, said candidate command, said search being implemented at least partially by said first equipment.

According to a second particular embodiment, said method comprises: a reception, at the level of the first equipment, of at least information representative of an audio sequence acquired by at least a second equipment connected to said network, and comprising at least one sequence textual deduced from said audio sequence; a search for an association of said textual sequence with at least one of said commands, said candidate command, said search being implemented at least partially by said first device. According to a particular characteristic, said search comprises an analysis of said information representative of said acquired audio sequence, to identify at least one audio element relating to said candidate command in connection with said acquired audio sequence. According to one particular characteristic, the method comprises generating at least one textual sequence from said audio sequence, and said search comprises an analysis of said generated textual sequence to identify at least textual element relating to said candidate command in connection with said generated text sequence. According to a particular embodiment, said generation comprises an assignment of a probability coefficient to said generated textual sequence and said search takes into account said probability coefficient.

According to a particular characteristic, said generation is implemented at least partially by said second equipment. According to a particular embodiment, said generation is implemented at least partially by at least one third piece of equipment and the method comprises, prior to said generation, a supply of at least one access element to said audio sequence to said third piece of equipment . This provision can in particular be implemented in the first or the second equipment. The generation can also be implemented in the first equipment, following the reception of said representative information by the first equipment.

According to a particular embodiment, the method further comprises a validation of said candidate command, and said candidate command is executed by said first equipment only when said validation is positive. According to a particular characteristic, said validation takes account of at least one selection criterion belonging to the group comprising: a state of said first equipment; an identifier of said second equipment; a type of said second equipment; a location information of said second equipment; a time range of acquisition of said candidate command; an identifier of a user at the origin of said audio sequence and / or designated in said audio sequence; a location information of a user at the origin of said audio sequence and / or designated in said audio sequence; a command already executed or being executed by said first equipment; an audio sequence acquired during a time span having an overlap with an acquisition time range of at least one audio sequence in which said candidate command has been found; a coefficient of relevance associated with said candidate command. According to a particular characteristic, said acquired audio sequence results from an acquisition of an audio content comprising: a measurement of an ambient sound level; a comparison of said measured sound level with a given first acquisition threshold; storing an audio sequence when said measured sound level exceeds said given first acquisition threshold. According to a particular characteristic, said storage is performed as long as said measured sound level is greater than a given second acquisition threshold.

According to a particular characteristic, said storage is performed as long as said measured sound level remains below the second acquisition threshold for a duration less than a given attenuation time. According to a particular characteristic, said storage is implemented during a given storage period.

According to a particular feature, the method implements a communication technique between said equipment of said network called "Publisher / Subscriber". According to another aspect, the invention relates to equipment, connected to a communication network and controllable by at least one command from said network. In particular, in certain embodiments, said equipment is configured to perform, at least partially, a search for an association of at least one piece of information representative of an audio sequence, acquired by at least a second piece of equipment connected to said network, with at least one of said commands, said candidate command. According to yet another aspect, the invention relates to so-called intermediate equipment, connected to a communication network to which is also connected a device 20 controllable by at least one command given from said network. In particular, in certain embodiments, said intermediate equipment is configured to perform, at least partially, a search for an association of at least one piece of information representative of an audio sequence, acquired by at least a second piece of equipment connected to said network. with at least one command, called the candidate command, of said controllable equipment. According to yet another aspect, the invention relates to a control system of a first equipment connected to a communication network and controllable by at least one command. In particular, in some embodiments, said system includes at least a second network-connected equipment comprising a processor configured to acquire at least one audio sequence, and said first equipment includes a processor configured to perform an association search. at least one piece of information representative of said audio sequence acquired with at least one of said commands, said candidate command.

Although not explicitly described, the embodiments presented can be implemented in any combination or sub-combination. For example, an embodiment in which the search includes an analysis of the acquired audio sequence can be combined with an embodiment in which the method comprises generating one or more text sequences from the acquired audio sequence, each being assigned a probability coefficient, and wherein the search includes an analysis of at least some of the textual sequences, in particular according to their probability coefficient. Other embodiments, easily conceivable by a person skilled in the art upon reading the present description, are also included in the scope of the invention. In particular, the invention is not limited to a single equipment acquisition of an audio sequence, or the control of a single equipment. According to another aspect, the invention relates to a computer program product comprising program code instructions for executing, the above control method, in any of the aforementioned embodiments, when said program is executed by a computer. In yet another aspect, the invention relates to a computer-readable storage medium on which is recorded a computer program comprising program code instructions for executing the control method of the invention, in any one of the following: aforementioned embodiments, when said program is executed by a computer. Such a computer readable storage medium may be in the form of a computer program product loaded onto at least one computer readable storage medium including computer readable program code instructions executable by a computer. Thus, in the present application, a computer-readable storage medium is considered to be a non-transitory storage medium having the intrinsic capacity to store information as well as the intrinsic ability to allow retrieval of the information it stores. A computer-readable storage medium may be for example, but not only, a system, apparatus or electronic, magnetic, optical, electromagnetic, infra-red, semiconductor equipment, or implementing a combination of techniques. previously mentioned. It should be noted that the following elements, which provide more specific examples of computer-readable storage media to which the principles of the invention may apply, are essentially cited for illustrative purposes and in no way constitute an exhaustive list. , as it will be easily interpreted by a person skilled in the art: a portable diskette for a computer, a hardware disk, a ROM-type memory (for Read Only Memory according to the English terminology), an erasable memory of the EPROM type (for Erasable Programmable Read Only). Memory according to the English terminology) or flash memory, a compact portable disk comprising a ROM (or CD ROM according to the English terminology), an optical storage equipment, a magnetic storage equipment, or any combination adapted from the previous elements. As would be readily apparent to one skilled in the art, according to variants, the aspects of the present invention may be implemented by a computer readable equipment, method or storage medium. Also, aspects of the present invention may be implemented in some embodiments as fully hardware components (e.g. an electronic component or an electronic card with components), or as fully software components (including for example components firmwares, software called "resident", microcode, etc ...). Other embodiments may implement both hardware components and software components. In addition, these hardware or software components can be implemented in the same physical equipment. The term "module" will generally be used herein to refer to a component that may correspond to a hardware component or a software component. In addition, aspects of the present invention may be implemented as a computer readable storage medium. Any combination of one or more computer readable storage media may be used. Thus, the invention may allow a voice command of a device, when it has no voice interface, through at least one other device. In addition, in certain particular embodiments, the invention may not require any coupling between the devices, in particular because of the "publisher / subscriber" type of mechanisms (or "Publisher / Subscriber" according to the English terminology) implemented.

Thus, the invention may allow in some embodiments a command by a third party equipment, without coupling with the equipment to be controlled. In particular, the third-party equipment does not need to know the services offered by the equipment to be controlled or its interface.

An equipment can in particular be controlled by a user in several ways, for example at the same time: directly, by means of a local control means to the equipment (for example a button or a touch screen located on its facade) ; remotely through a remote control accessory, indirectly, through at least one other equipment, according to one embodiment of the invention, by the combination of a voice command, acquired through other equipment to a command of the equipment to be controlled. Therefore, it may be possible for a user to choose the control means that he wishes to use to act on a device to be controlled and in particular to opt for indirect control when the equipment does not have an audio interface or is located away from the user (for example when the user wishes to order equipment from his place of work from his place of work) or simply when the user is out of reach of the audio interface of the equipment (because the user and the equipment to be controlled are in different rooms of the user's home for example). In addition, certain embodiments, for example embodiments implementing an automatic acquisition speech interface, allow the user to control a device without having to use any accessory (such as a remote control) or even to using one's hands, which can be practical when the user is unable to use his hands, for example because of a disability, or because it performs a parallel manual activity. The invention can also make it possible to avoid equipping an object with an expensive user interface (such as a touch screen for example) and / or poorly adapted to a small object (such as a connected watch, for example ). It is therefore particularly suited to an objective of reducing production costs or the size of a communicating object. 4. List of figures. The invention will be better understood, and other features and advantages will appear on reading the detailed description which follows, relating to a particular embodiment, the description referring to the accompanying drawings in which: - Figures 1A, 1B and 1C each show a synoptic view of a system in a first particular embodiment of the invention; - Figure 2 is a block diagram illustrating the control method of the invention, in a particular embodiment, compatible with the embodiments described in connection with Figures 1A, 1B and 1C. - Figures 3 to 5 respectively show an example of hardware architecture of an intermediate equipment, an acquisition equipment and a control equipment according to a particular embodiment of the invention. The same element is designated in the set of figures by the same reference sign. The figures presented are for illustrative purposes only and in no way limit the invention to the illustrated embodiments. 5. Description of embodiments of the invention. 5.1 General principle The general principle of the invention is to allow indirect voice control of a device (called "equipment to be controlled"), for example because it does not have a voice interface, or because it found at a great distance from a user, using at least one third-party equipment, for example equipment located near the user and / or having audio acquisition means of longer range than the equipment to be controlled and comprising means of communication with the equipment to be controlled, for example means of communication through a home network or a broadband network such as an internet-type network. 5.2. PRESENTATION OF A PARTICULAR EMBODIMENT Now, in connection with FIG. 1A, a particular embodiment of the invention is presented for a system 100 of a communication network, for example a home network such as a WIFI network or a wired network.

In the particular embodiment illustrated, the system 100 comprises a device 110 comprising means for acquiring (for example a microphone) audio data 112,. It also includes a control equipment 120, for example a set-top box. In the embodiment illustrated in FIG. 1A, the audio acquisition equipment 110 does not know how to analyze an audio sequence that it acquires and essentially comprises means for acquiring, storing and transmitting at least one audio sequence. 112 through the communication bus 140. The equipment to be controlled 120 comprises means for obtaining an audio sequence 112 from the information conveyed on the communication bus 140. It comprises means for searching a connection association. the audio sequence obtained 112, with one of the commands he can interpret. FIG. 1B is a variant of the embodiment illustrated in FIG. 1A in which the system 100 furthermore comprises an intermediate equipment 130. It may for example be a LAN equipment remaining continuously active, notably a gateway allowing a interconnection of the local network with a remote network, or a dedicated interconnection equipment, for example of low consumption. In the embodiment illustrated in FIG. 1B, the acquisition equipment 110 is for example identical to the acquisition equipment of FIG. 1A and the intermediate equipment 130 comprises means for obtaining an audio sequence 112 from the information conveyed on the communication bus 140. In certain embodiments, the intermediate equipment 130 may furthermore comprise means for searching, from the audio sequence obtained, for a command that can be interpreted by the equipment to be used. 120, and transmission means of this command on the communication bus 140. Such an embodiment may for example be transparent vis-à-vis the equipment to be controlled and therefore allow compatibility with existing control equipment to control of the market. In other embodiments, as in the embodiment of FIG. 1B, the intermediate equipment 130 may comprise means for generating a textual sequence 122, for example voice recognition and transmission means. this textual sequence 122 on the communication bus 140. In the embodiment of FIG. 1B, the intermediate equipment is devoid of means for analyzing this textual sequence. In such an embodiment, the equipment to be controlled 120 obtains a textual sequence 122 conveyed on the communication bus 140. It comprises search means, from the conveyed textual sequence, representative of the acquired audio sequence, of a command that can be interpreted by the equipment to be controlled 120. Such an embodiment may allow control of a piece of equipment without the intermediate equipment being aware of the commands that can be interpreted by the equipment to be controlled. Therefore, such an embodiment can allow the control of several equipment to be controlled by the same intermediate equipment and an addition of a device to be controlled or an evolution of its interface in a transparent manner for the intermediate equipment. It should be noted that in the embodiment of FIG. 1B, the equipment to be controlled can also receive audio sequences from the acquisition equipment and analyze both audio sequences and text sequences. Such an embodiment offers the advantage of allowing an adaptation of the method to the periods of unavailability of the intermediate equipment (if it implements the generation of textual sequences). It can also allow a more complete search from an acquired or more secure audio sequence (for example when it is a critical command that will have to be identified both in the audio sequence and in the generated text sequence. from the audio sequence). As explained above, the intermediate equipment 130 is optional. Indeed, in some embodiments, the acquisition equipment 110 or the equipment to be controlled 120 may themselves comprise search and association means of a given command from at least one information item. representative of an audio sequence (for example a part of the audio sequence itself, an access element (like a URL) to the audio sequence and / or a textual sequence generated from the acquired audio sequence) and adequate means of communication.

In the embodiment illustrated in FIG. 1C, the acquisition equipment 110 comprises means for generating a textual sequence 122 from the acquired audio sequence, for example voice recognition means, and transmission means of this textual sequence 122 on the communication bus 140. The equipment to be controlled 120 of FIG. 1C may for example be identical to the equipment to be controlled 120 of FIG. 1B. Again, as explained in connection with FIG. 1B, the equipment to be controlled 120 can receive both audio sequences and text sequences from the acquisition equipment 110 and analyze these two types of sequences, to perform a more complete or more secure search.

Figures 1A, 1B and AC illustrate deliberately simple embodiments of the invention, to better highlight some of the concepts of the invention. It is clear, however, that, without departing from the scope of the invention, the system may include several acquisition equipment (for example several microphones located at various locations), several intermediate equipment and / or more equipment to control. In particular, intermediate equipment may be necessary for certain voice commands of the user and not for others, depending on the degree of improvement of the acquisition equipment present in the vicinity of the user (for example it may be a simple independent microphone, or a tablet, a smart phone or a personal computer equipped with a microphone) or equipment to control (for example a thermostat or a set-top-box). In the particular cases illustrated in FIGS. 1A, 1B and 1C, the acquisition equipment 110, to be controlled 120 and intermediate 130 communicate via a data bus 140. The exchanges between the equipment of the system 100 may notably be based on an interoperable standard data exchange protocol, for example the "DDS" protocol (for "Data Distribution Service" in English terminology) managed by OMG (for "Object Management Group"). Thus, in the particular embodiment presented, the equipment belongs to a platform, such as the "Qeo" platform of the requester, based on a distributed architecture and using, for example, an "editor" type data exchange model. subscriber "(or" Publisher / Subscriber "in the English terminology) implemented in the" DDS "protocol. Such a platform offers the advantage of enabling point-to-point or multipoint communication of equipment without requiring mutual knowledge. In one variant, the exchanges between equipments can use a "Mosquitto" type platform, based on a centralized architecture, in which the exchanges between equipments are implemented via an intermediary server, which manages the subscriptions and the notifications of the equipments of the network. , for example an architecture using the "MQTT" protocol (for MQ Telemetry Transport).

In an embodiment using a DDS or MQTT protocol, the publications take the form of messages, associated with a given type of content (hereinafter, we will speak for the sake of simplification of "type of message"). Each message corresponds to a data structure, including in particular a content and / or an element of access to this content, and a type of content. The data structure may also include other elements, which vary according to the type of content of the message. The subscribing equipment subscribes by providing the list of message types that interest them. For example, in the embodiment illustrated in FIG. 1B, the intermediate equipment 130 can subscribe to receive the published messages of the audio type and the equipment to be controlled 120 can subscribe to receive messages of the "text" type. "," Phrase ", or" user message ". In connection with FIG. 2, the dynamic operation of the method of the invention is presented in a particular embodiment compatible with the embodiment illustrated in FIG. 1B.

In the embodiment of FIG. 2, the method comprises an acquisition 200 of an audio sequence, by the acquisition equipment 110. This acquisition 200 can for example be triggered and / or stopped by an action of the user. for example by pressing an "on" key and / or "stopping the acquisition equipment. In other embodiments, audio acquisition 200 may be automatically controlled. For example, in the illustrated embodiment, the acquisition 200 includes an evaluation of an ambient sound level 202, and a comparison 204 with a reference sound level. The storage of an audio sequence 206 is for example triggered by the passage of the ambient sound level above a first reference sound level. It can be carried out for a given duration (so as to obtain audio sequences of constant size, for ease of analysis for example), and / or as long as the ambient sound level remains above a second sound level of reference. In this case, the method may comprise, parallel to the storage of the audio sequence, a measurement (not shown), for example a periodic measurement, of the sound level and a comparison 208 with a second reference level. The storage 206 may also continue as long as the sound level remains below the second threshold for a duration less than a given threshold duration. According to the embodiments, the first reference audio level (or acquisition trigger threshold), and the second reference audio level (or acquisition stop threshold) may be the same, or different. For example, the acquisition trigger threshold may be greater than the acquisition stop threshold, to take account of a distance and / or a change of position of the user while he gives his voice command. or a drop in tone, desired or otherwise, of the user for example. In the embodiment illustrated in FIGS. 1B and 2, which implement intermediate equipment 130, the method comprises a supply 210 to the intermediate equipment of at least one piece of information representative of the acquired audio sequence, and in particular of a access element to the audio sequence. In some embodiments, where the audio sequences are small (eg a short order "TV On", "TV off"), the access element may be the audio sequence itself.

In other embodiments, adapted in particular to larger audio contents, the access element may be location information of the acquired audio sequence. It can for example be, when the acquired audio sequence is stored as one or more audio files, one or more URLs and / or an identifier of a device, giving the location of audio files. These files may, for example, be stored locally in the acquisition equipment, for example in a flash memory, as in the embodiment presented, or remotely on a third-party device, for example a piece of equipment on the local network or a remote server. . In this case, the method comprises, before the publication of the acquired audio sequence, a transmission (not shown) of the audio file to the third-party equipment where it will be stored. More specifically, in an embodiment compatible with the embodiment illustrated in FIG. 1, where the platform to which the equipment belongs is a platform using the distributed "DDS" protocol, after the acquisition of an audio sequence 200, the provision 210 comprises a publication of an "audio" type message (that is, as explained above, a data structure containing at least one piece of information representative of the audio sequence, in particular an access element to the audio sequence), of which the intermediate equipment 130, in particular, is notified. In the particular embodiment presented, the information representative of the acquired audio sequence may in particular comprise, in addition to the access element to the acquired sequence, at least one piece of information belonging to the group comprising: a start and / or end of capture of the audio sequence; an identifier of at least one acquisition equipment; location information of at least one acquisition equipment; a configuration parameter of the acquisition equipment (default language, acquisition sound level threshold, etc.); a type of acquisition equipment (isolated microphone, computer equipped with a microphone, etc.); an identifier of at least one user at the origin of the acquisition (that is to say at least one user who has pronounced the audio sequence).

In one embodiment where the audio sequence is not provided directly in the message sent by the acquisition equipment, the supply 210 of the information representative of the audio sequence also comprises obtaining the audio sequence acquired according to the element access provided. For example, in the embodiment illustrated in FIGS. 1B and 2, the method comprises reading the contents of a local URL to the acquisition equipment 110 by the intermediate equipment 130, to obtain the acquired audio sequence. The method also comprises a search 220, of an association of at least one of the information representative of the acquired audio sequence with at least one command of the equipment to be controlled 120. In other words, it means recognizing from the words spoken by the user a command to be taken into account by the equipment to be controlled. In certain embodiments, at least one element that can be associated with certain commands (for example in audio and / or textual form) can be defined.

For example, heating equipment can be controlled by the following set of commands: ON (T): Turning on the equipment at an indicated heating temperature; STOP (): Stop the equipment; INCREASE (): Increase heating temperature; LOWER (): Decrease heating temperature. Items associated with the ON (T) Start command may for example include keywords in at least one language, for example "Walk", "I'm cold", "temperature", and "degree and words" keys forming probable numbers, vis-à-vis the use of equipment (such as "twenty-two" for heating equipment, but not "thousand" for example). The elements associated with STOP () stop command may for example include the keywords "Stop" and "Stop", in sound or text form. Items that are associated with the INCREASE (Augmentation) command may for example include the items "Warmer", "I'm cold", "Increase" and 20 "Heating". Items that can be associated with the LOWER () "Decrease" command may include, for example, the "Less Hot", "I'm Too Hot", "Decrease" and "Heating" items. In particular, some elements can be associated with several different commands (for example the keyword "Heating" in the example above). In some embodiments, the search 220 may include in particular an analysis 226 of the audio sequence acquired to identify at least one audio element that can be associated with the control of the equipment 120. This search may also include a generation 222 of at least a textual sequence, from the audio sequence and then an analysis of the textual sequences generated to identify at least one textual element associable with the control of the equipment 120. According to the embodiments, such a textual analysis can conclude that identification of a textual element as sociable when it is found integrally in a generated sequence, surrounded or not by characters delimiting words, such as spaces or punctuation marks, or when only part of the textual element is found in the sequence (for example, if the element is a word, a root of that word). The generated text sequences may for example be conversions of at least a portion of the acquired audio sequence, performed locally by the intermediate equipment 130, when it comprises voice recognition means, or by means of a remote application , for example an internet application of the "Google Speech Web Service" type, having as input parameter an audio content and as an output parameter a textual content.

In the illustrated embodiment, the conversion of an audio sequence can lead to the generation of several textual sequences, more or less probable (for example when a sound can be translated into several phonetically close words, during voice recognition, but more or less probable according to the words already recognized). For example, it can be an ordered list of textual sequences, depending on their probability. It can also be a set of textual sequences each assigned a probability coefficient. In a particular embodiment where several acquisition equipments are used and where the same intermediate equipment can receive several audio sequences corresponding to the same audio event, the probability coefficient of a textual sequence can be increased if it is provided several times relatively audio sequences having acquisition time ranges having at least one overlap. In such an embodiment, the method may for example comprise a storage of the generated textual sequences and a delay before the transmission (see below) of the generated sequences to refine the probability coefficient of a sequence.

When the search was successful and made it possible to find a specific command (called "candidate command") associated with the audio sequence, for example because the analysis made it possible to identify at least one textual or audio element that can be associated with the command candidate, a relevance coefficient can be assigned to a candidate command. This coefficient can in particular take into account the probability of a textual sequence in which an element that can be associated with the candidate command has been found, a quality of acquisition of an equipment for acquiring an audio sequence at the same time. origin of the candidate command, of a number of elements associated with the candidate command identified in at least one generated text sequence or in the audio sequence, predefined rules of combination of certain elements as sociable with the found identified command, of a probability of identifying an element that can be associated with the identified candidate command (and calculated, for example, to take into account a number of textual sequences associated with the same audio sequence containing this associable element). The relevance coefficient may also take into account the associable elements identified in several audio sequences relating to the same audio event and having acquisition time ranges having at least one overlap, similar to what is explained above in the particular case. acquisition by a single acquisition equipment. In the particular embodiment illustrated in FIGS. 1B and 2, where the search 220 is performed by the intermediate equipment, when the search has enabled the identification of at least one candidate command, the method comprises a transmission 228 to the first equipment information representative of at least one of the candidate commands. This is for example a publication of a message, type "text" for example, containing at least one of the candidate commands to the equipment to be controlled.

According to the embodiments, the transmission 228 may relate to one or more candidate commands: for example it may relate to a single candidate command (the commands being transmitted successively), or to all the candidate commands, or to the candidate command considered the most relevant according to its relevance coefficient (for example because it has the greatest or the smallest relevance coefficient, if it concerns numerical values), or to each of the n most relevant candidate commands (n greater than 1), or candidate commands with a relevance coefficient greater than or less than a certain relevance threshold. According to the embodiments, the transmission 228 may comprise the commands themselves or a link to their location (local to the intermediate or remote equipment). The transmission 228 may furthermore comprise additional information, in addition to the candidate command or commands, in particular complementary information derived from or extracted from the representative information relating to an audio sequence provided (see supply 210 above) by at least one piece of equipment. acquisition, for example at least one item of information belonging to the group comprising: an item associated with an identified candidate command, at least part of an audio or textual acquisition-related sequence (for example to allow the equipment to be controlled to find in one of these sequences parameter values necessary for certain candidate commands); a coefficient of relevance of a candidate command; a timestamp of beginning and / or end of acquisition of an acquired audio sequence; a degree of accuracy of said time stamp; an identifier of an acquisition equipment of an acquired audio sequence; a type of acquisition equipment (microphone, ...) of an acquired audio sequence; a quality of acquisition of an acquired audio sequence; an identifier of the user having pronounced an acquired audio sequence.

For example, the identifier of a user can be determined during speech recognition by comparison with pre-recorded voice samples associated with particular user identifiers. It can also be provided by the acquisition element when it comprises means for identifying a user (for example by the user's "login").

In the embodiment described in FIG. 2, the method also comprises a validation 230 of a candidate command by the first equipment. This validation can take into account additional information transmitted with the candidate command. For example, the user's identifier can be used to check, in association with a profile of the user, his right to execute a transmitted candidate command. A received command can in particular be ignored if it is impossible or not coherent (because of the state of the equipment or one of its peripherals), or if another command relating to the same audio event has already been received. or executed or running (that is, if both commands are for audio sequences whose acquisition range has time overlay).

Such an embodiment, by limiting the association of an acquired audio sequence to a single command of the equipment to be controlled, offers the advantage of avoiding multiple entries of the same command, which could have a unwanted cumulative effect. For example, an "increase sound" command will be processed once. FIGS. 3 and 4 show an example embodiment of equipment 130, 110 and 120 of FIG. 1B in an embodiment compatible with the particular embodiment described in FIG. that the word "register" used in the description of a memory designates in each of the memories mentioned, as well a memory area of low capacity (some binary data) that a large memory area (for storing an entire program or all or part of audio or textual sequences). An embodiment of the intermediate equipment 130 of FIG. 1B is now shown in FIG.

The intermediate equipment 130 corresponds for example to a laptop (from the English "laptop"), a tablet or a smartphone (English Smartphone) or a gateway to access a local network. In the particular embodiment illustrated, the intermediate equipment 130 comprises the following elements, interconnected by an address and data bus 300 which also carries a clock signal: a microprocessor 31 (or CPU); one or more devices 1/0 (English "Input / Output" or in French "Input / Output") 34, such as for example LEDs, a keyboard, control buttons etc. ; a non-volatile memory type ROM (Read Only Memory) 35; Random Access Memory (RAM) 36; a communication interface 37 configured for receiving data via a wireless connection (for example of the WIFI® or Bluetooth type) or wired connection (for example of the Ethernet type); a communication interface 38 configured for the transmission of data via a wireless connection (for example of the WIFI® or Bluetooth type) or wired connection (for example of the Ethernet type); 39. The intermediate equipment 130 may also comprise, in certain embodiments, as shown in FIG. 3, a graphics card 32 and a display-screen display device 33 connected directly to the graphics card 32 by a Dedicated bus 330. According to one variant, the terminal comprises several microprocessors. According to another variant, the power supply 39 is external to the intermediate equipment 130.

In the particular embodiment illustrated, the intermediate equipment 130 also comprises means for subscribing to "audio" type messages and for receiving such messages, as well as means for publishing "text" type messages and / or "User message". It further comprises means for generating at least one textual sequence from an audio sequence.

On power-up, the microprocessor 31 loads and executes the instructions of the program contained in a register 360 of the RAM 36, and in particular the algorithms implementing certain steps of the method specific to the invention and described above. In the embodiment illustrated in FIG. 3, the microprocessor 31 may in particular be configured to search for an association of at least one piece of information representative of an audio sequence, acquired by at least a second piece of equipment connected to said network, for example a audio sequence designated by an "audio" type message received by said communication interface 37, with at least one given command, relating to a control device connected to a communication network to which the intermediate equipment is also connected. FIG. 4 shows an example of a hardware embodiment of the acquisition equipment 110 of FIG. 1B. In the particular embodiment illustrated, the acquisition equipment 110 comprises the following elements, interconnected by an address and data bus 400 which also carries a clock signal: a microprocessor 41 (or CPU) ; one or more devices 1/0 (English "Input / Output" or in English "Input / Output") 44, including a microphone, LEDs; a non-volatile memory type ROM (Read Only Memory) 45; RAM or Random Access Memory (RAM) 46; a communication interface 48 configured for the transmission of data, in particular of the audio type, via a wireless connection (for example of the WIFI® or Bluetooth type) or wired connection (for example of the Ethernet type); In the particular embodiment illustrated, the acquisition equipment 110 also comprises means for publishing "audio" type messages. In certain particular embodiments, as illustrated for example in FIG. 4, the acquisition equipment 110 may also comprise a communication interface 47 configured for receiving data (for example subscription requests from the data carrier). intermediate equipment 130 or equipment to be controlled 120) via a wireless connection (for example of the WIFI® or Bluetooth type) or wired connection (for example of the Ethernet type). According to one variant, the acquisition equipment 110 comprises several microprocessors.

According to another variant, the power supply 49 is external to the acquisition equipment 110. At power-up, the microprocessor 41 loads and executes the instructions of the program contained in a register 460 of the RAM 46, and in particular the algorithms implementing certain steps of the method specific to the invention and described above. In the embodiment illustrated in FIG. 4, the microprocessor 41 may in particular be configured to acquire an audio sequence, store it in a register of the RAM 46 and transmit information relating to the acquisition of the audio sequence on the interface. of communication 48.

An embodiment of the equipment to be controlled 120 of FIG. 1B is now shown in FIG. The equipment to be controlled 120 corresponds, for example, to a set-top box or to an intelligent household appliance equipment connected to a local area network. In the particular embodiment illustrated, the equipment to be controlled 120 comprises the following elements, interconnected by an address and data bus 500 which also carries a clock signal: a microprocessor 51 (or CPU); one or more devices 1/0 (of the English "Input / Output" or in French "Input / Output") 54, such as for example a keyboard, light-emitting diodes (or LEDs for "Light-Emitting Diode" according to English terminology), etc. ; a non-volatile memory of the ROM type (of the English "Read Only Memory") 55; a random access memory (Random Access Memory) 56; a communication interface 57 configured for receiving data via a wireless connection (for example of the WIFI® or Bluetooth type), in particular data of the audio or textual type; a power supply 59. The equipment to be controlled 120 may optionally also comprise, as represented in FIG. 5, a graphics card 52 and a display device 53 of the display screen type directly connected to the graphics card 52 by a dedicated bus 500. In the particular embodiment illustrated, the equipment to be controlled 120 also comprises means for subscribing to messages of the "text" and / or "user message" and / or "audio" type and for receiving such messages. and means for executing at least one command designated by such a message.

In particular, in the embodiment of FIG. 5, the equipment to be controlled comprises an interface a communication interface 58 configured for the transmission of data via a wireless connection (for example of the WIFI® or Bluetooth type), (by example subscription requests to the intermediate equipment 130 or the acquisition equipment 110) via a wireless connection (for example WIFI® or Bluetooth type) or wired (for example Ethernet type). According to one variant, the terminal comprises several microprocessors. According to another variant, the power supply 59 is external to the device 120. At power-up, the microprocessor 51 loads and executes the instructions of the program contained in a register 560 of the RAM 56, and in particular the algorithms implementing certain process steps specific to the invention and described above. In the embodiment illustrated in FIG. 5, the microprocessor 51 may in particular be configured to search for an association of at least one piece of information representative of an audio sequence, acquired by at least a second piece of equipment connected to said network, for example a audio sequence designated by a message (of the "audio" type or of the "text" type, for example) received by said communication interface 57, with at least one given command, relative to the equipment to be controlled, and / or to validate a candidate command received by the communication interface 57, the candidate command being executed by the equipment to be controlled only when the validation is positive.

Claims (18)

REVENDICATIONS1. A method of controlling a first device connected to a communication network and controllable by at least one given command, said method being characterized in that it comprises: a search for an association of at least one piece of information representative of an audio sequence, acquired by at least a second equipment connected to said network, with at least one of said commands, called the candidate command. 2. Control method according to claim 1 characterized in that said search comprises an analysis of said information representative of said acquired audio sequence, to identify at least one audio element relating to said candidate command in connection with said audio sequence acquired. 3. Control method according to one of claims 1 or 2 characterized in that the method comprises generating at least one textual sequence from said audio sequence, and in that said search includes an analysis of said textual sequence generated to identify at least one textual item relating to said candidate command in conjunction with said generated textual sequence. 4. Control method according to claim 3 characterized in that said generation comprises an assignment of a probability coefficient to said generated textual sequence and in that said search takes into account said probability coefficient. 5. Control method according to claims 3 or 4, characterized in that said generation is implemented at least partially by said second equipment. 6. Control method according to any one of claims 3 to 5, characterized in that said generation is implemented at least partially by at least a third equipment and in that the method comprises, prior to said generation, a supply at least one access element to said audio sequence to said third equipment. 7. Control method according to claim 1 to 6, characterized in that said search is implemented at least partially by said first equipment. 8. Control method according to claim 1 to 7 characterized in that it further comprises a validation of said candidate command, and in that said candidate command is executed by said first equipment only when said validation is positive. 9. Control method according to claim 8 characterized in that said validation takes into account at least one selection criterion belonging to the group comprising: a state of said first equipment; An identifier of said second equipment; a type of said second equipment; a location information of said second equipment; a time range of acquisition of said candidate command; an identifier of a user at the origin of said audio sequence and / or designated in said audio sequence; a location information of a user at the origin of said audio sequence and / or designated in said audio sequence; a command already executed or being executed by said first equipment; an audio sequence acquired during a time span having an overlap with an acquisition time range of at least one audio sequence in which said candidate command has been found; a coefficient of relevance associated with said candidate command. 10. Control method according to any one of claims 1 to 9, said method being characterized in that said audio sequence acquired results from an acquisition of an audio content comprising: a measurement of an ambient sound level; a comparison of said measured sound level with a given first acquisition threshold; storage of an audio sequence when said measured sound level exceeds said given first acquisition threshold. 11. Control method according to claim 10 characterized in that said storage is performed as said measured sound level is greater than a second acquisition threshold given. 12. Control method according to claim 11 characterized in that said storage is performed as said measured sound level remains below the second acquisition threshold for a period less than a given duration of attenuation. 13. Control method according to any one of claims 10 to 12 characterized in that said storage is implemented for a given storage period. 14. Control method according to one of claims 1 to 13, characterized in that it implements a communication technique between said equipment said network called "Publisher / Subscriber". Equipment connected to a communication network and controllable by at least one command from said network, said equipment being characterized in that it is configured to perform, at least partially, a search for an association of at least information representative of an audio sequence, acquired by at least a second equipment connected to said network, with at least one of said commands, said candidate command. 16. A control system of a first device connected to a communication network and controllable by at least one given command, said system being characterized in that it comprises at least a second equipment connected to the network comprising a processor configured to acquire at least one minus an audio sequence; and in that said first device comprises a processor configured to perform, at least partially, a search for an association of at least one piece of information representative of said audio sequence acquired with at least one of said commands, said candidate command. 17. Computer program product, characterized in that it comprises program code instructions for executing, the control method according to any one of claims 1 to 14, when said program is executed by a computer. A computer-readable recording medium on which is recorded a computer program comprising program code instructions for executing the control method of the invention, according to any one of claims 1 to 14, when said program is run by a computer.25