CN111161731A - Intelligent off-line voice control device for household electrical appliances - Google Patents
Intelligent off-line voice control device for household electrical appliances Download PDFInfo
- Publication number
- CN111161731A CN111161731A CN201911395573.8A CN201911395573A CN111161731A CN 111161731 A CN111161731 A CN 111161731A CN 201911395573 A CN201911395573 A CN 201911395573A CN 111161731 A CN111161731 A CN 111161731A
- Authority
- CN
- China
- Prior art keywords
- voice
- control
- control function
- function item
- module
- 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.)
- Pending
Links
- 238000013500 data storage Methods 0.000 claims description 57
- 238000012545 processing Methods 0.000 claims description 57
- 239000013598 vector Substances 0.000 claims description 48
- 230000002708 enhancing effect Effects 0.000 abstract 1
- 238000000034 method Methods 0.000 description 16
- 230000008569 process Effects 0.000 description 9
- 238000010276 construction Methods 0.000 description 8
- 230000004048 modification Effects 0.000 description 8
- 238000012986 modification Methods 0.000 description 8
- 230000005236 sound signal Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 238000004378 air conditioning Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001172 regenerating effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L2015/088—Word spotting
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Selective Calling Equipment (AREA)
Abstract
The embodiment of the invention discloses an intelligent household appliance off-line voice control device, which locally stores a control instruction packet comprising at least one control function item and at least two control instruction words which have corresponding relations with each control function item, acquires voice instruction information of a recognition user to obtain a voice recognition result, searches for the control instruction words matched with the voice recognition result, locally searches for the control function items corresponding to the control instruction words according to the control instruction words, and controls an intelligent household appliance to execute the control function items. The embodiment of the invention realizes that a plurality of control command words correspond to the control function items in the off-line mode, and realizes that a plurality of control command words control one control function item, thereby improving the effective rate of voice control and enhancing the control experience of users on intelligent household appliances.
Description
Technical Field
The invention relates to the field of intelligent household appliances, in particular to an intelligent household appliance offline voice control device.
Background
With the continuous development and progress of computer technology and big data processing technology, the voice technology is rapidly developed, and the voice recognition technology is widely applied to the control function of intelligent household appliances. The intelligent household appliance with the voice control function can be controlled by sending a voice command by a user, for example, the user can control the refrigerator to open the door and adjust the temperature and the mode of the refrigerator through the voice command.
At present, voice control is carried out on intelligent household appliances, most of off-line control command words are preset in the local equipment by household appliance manufacturers, and voice control instructions correspond to household appliance control functions one to one. Only when the voice instruction spoken by the user is the same as the preset control command word, the device can judge that the voice recognition is successful, thereby executing the device control function corresponding to the voice instruction.
However, in the process of actually using the voice-controlled intelligent household appliance, the user often cannot remember the preset control command words of the household appliance, or the voice control is directed to different users, and the users have different speaking habits and modes, and the preset control command words do not accord with the ordinary speaking habits of the users. The user cannot timely speak out the control instruction or speak out a wrong control instruction in the process of using the voice to control the intelligent household appliance, so that the voice recognition process is overtime, the voice recognition is wrong, and the voice recognition is invalid. Therefore, the effective rate of voice control is reduced, and the control experience of a user on the intelligent household appliance is influenced.
Disclosure of Invention
The embodiment of the invention provides an intelligent household appliance offline voice control device, which can improve the effective rate of voice control and enhance the control experience of a user on an intelligent household appliance.
The embodiment of the invention adopts the following technical scheme:
an intelligent household electrical appliance off-line voice control device, which is used for intelligent household electrical appliance off-line voice control, and comprises: the voice recognition system comprises a data storage module, a voice acquisition module, a voice processing module, a data processing module and a control module;
the data storage module is arranged locally and used for storing a control instruction packet locally, and the control instruction packet comprises at least one control function item and at least two control command words which correspond to the control function items;
the voice acquisition module is used for acquiring voice instruction information of a user when the intelligent household appliance enters a voice acquisition mode and transmitting the voice instruction information to the voice processing module;
the voice processing module is used for recognizing the voice instruction information to obtain a voice recognition result;
the data processing module is used for searching a first control command word matched with the voice recognition result;
the control module is used for searching a first control function item corresponding to the first control command word in the data storage module according to the first control command word, and controlling the intelligent household appliance to execute the first control function item.
Optionally, the data storage module is further configured to update the control instruction packet according to a control command word edited by a user and corresponding to a second control function item, where the second control function item is any control function item in the at least one control function item.
Optionally, the method further includes: a display module;
the display module is used for controlling a display page to display at least one control function item contained in the control instruction packet and at least two control command words which have corresponding relations with each control function item;
the display module is further used for displaying a display page for editing the second control function item to a user, sending a prompt for editing the second control function item, and acquiring a control command word corresponding to the second control function item edited by the user through the display page; or when the display page displays at least one control function item contained in the control instruction packet and at least two control command words having a corresponding relation with each control function item, a prompt of the second control function item through voice editing is displayed for a user, and the voice acquisition module is further used for receiving the control command words corresponding to the second control function item through voice editing by the user.
Optionally, the data processing module is further configured to obtain, through the voice acquisition module, a control command word edited by the user in a voice manner, where the control command word corresponds to the second control function item.
Optionally, the data storage module is further configured to, when the data processing module does not find the first control command word matching the voice recognition result, extract the control command word from the voice recognition result, correspond the extracted control command word to a third control function item in the at least one control function item according to a user instruction, and update the control instruction packet.
Optionally, the data storage module is further configured to obtain a command word grammar file for calibrating the command word combination rule, and determine a speech recognition model file related to the command word grammar file, where the speech recognition model file includes standard speech feature vectors of all control command words in the control instruction packet.
Optionally, the voice processing module is configured to convert voice instruction information into a voice feature vector, match the voice feature vector with a standard voice feature vector in the voice recognition model file, obtain the standard voice feature vector closest to the voice feature vector from the voice recognition model file, and determine the voice recognition result.
Optionally, the voice recognition result obtained by the voice processing module is a character;
and the data processing module is also used for searching the matched first control command word from the storage module according to the characters if the characters are contained in the command words of the storage unit and the command word grammar file.
Optionally, the data storage module is configured to assign an ID to each control function item; and establishing a storage table, wherein the ID of each control function item is correspondingly stored with all control command words which respectively have corresponding relations with each control function item in real time.
Optionally, the voice processing module is further configured to determine a voice recognition threshold;
the voice processing module is further configured to compare the voice recognition threshold with a preset threshold, where the voice recognition threshold is greater than the threshold, the voice recognition result is accurate, and the voice recognition threshold is less than the threshold, the voice recognition result is invalid.
The off-line voice control device for the intelligent household appliance based on the technical scheme acquires voice instruction information and voice instruction information of the recognition user to obtain a voice recognition result by locally storing the control instruction packet containing at least one control function item and at least two control instruction words corresponding to each control function item, searches the control instruction words matched with the voice recognition result, locally searches the control function items corresponding to the control instruction words according to the control instruction words, and controls the intelligent household appliance to execute the control function items, so that the control function items corresponding to a plurality of control instruction words are realized in an off-line mode, one control function item is controlled by a plurality of control instruction words, the voice control efficiency is improved, and the control experience of the user on the intelligent household appliance is enhanced.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
Fig. 1 is a schematic structural diagram of an intelligent offline voice control device for home appliances according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of another intelligent appliance offline voice control apparatus according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of another intelligent appliance offline voice control apparatus according to an embodiment of the present invention;
fig. 4 is a flowchart of a method for performing offline voice control on an intelligent home appliance by using an offline voice control device for an intelligent home appliance according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the invention, as detailed in the appended claims.
The intelligent household appliance offline voice control device can solve the technical problem that a user cannot speak a control instruction or speak an incorrect control instruction in time due to the fact that the user cannot remember a control instruction word preset by a household appliance or the preset control instruction word does not accord with the speaking habit of the user in the process of controlling the intelligent household appliance by using a voice instruction, and therefore voice control is invalid.
The intelligent household appliance offline voice control device is applied to intelligent household appliances, and the intelligent household appliances can be refrigerators, televisions, air conditioners and other devices.
In the embodiment of the present invention, the first control function item, the second control function item, and the third control function item are only divided for convenience of description, and the division is not limited to the embodiment of the present invention, and may be the same or different control function items in different scenarios.
Example 1
As shown in fig. 1, the present embodiment is an intelligent home appliance offline voice control apparatus, which is applied to an offline state (i.e. an unconnected network state) of an intelligent home appliance, and includes: the device comprises a data storage module 11, a voice acquisition module 12, a voice processing module 13, a data processing module 14 and a control module 15.
The data storage module 11 is disposed locally (for example, an intelligent household electrical appliance device), and is configured to store a control instruction packet locally, where the control instruction packet includes at least one control function item and at least two control command words having a corresponding relationship with each control function item.
In this embodiment, the control function item is a function that can be realized by the intelligent household electrical appliance, and different intelligent household electrical appliances may be different. For example, air conditioning to increase or decrease temperature, television to adjust volume, etc. In different embodiments, the control function item may be a program for implementing a control specific function, or may be a module for implementing a control specific function.
In this embodiment, the number of the control command words having the corresponding relationship in each control function item is at least two, and the at least two control command words may be set according to different user habits or different expression modes, so that the success rate of matching the control command words with the control function items may be increased, and the user experience may be enhanced. When storing the control function item and at least two control command words having a correspondence relationship therewith, a correspondence relationship table may be established in which the control function item, a plurality of (at least two) control command words corresponding to the control function item, and a correspondence relationship between the control function item and the plurality of control command words are recorded.
The voice acquisition module 12 is configured to acquire voice instruction information of the user when the intelligent household appliance enters a voice acquisition mode, and transmit the voice instruction information to the voice processing module 13.
The voice acquisition mode of the intelligent household appliance can be started according to the instruction of the user, and the voice acquisition module 12 acquires the voice input instruction information of the user after the intelligent household appliance enters the voice acquisition mode.
In this embodiment, the voice collecting module 12 may be disposed on an external panel of the intelligent home appliance or around the user operation interface, so as to collect the voice of the user. The voice capture module 12 may be provided with a manual button switch and a work indicator light to facilitate the user to manually turn on and off and understand the working status of the voice capture module 12.
In this embodiment, the voice collecting module 12 may be one or more microphones, and the plurality of microphones may be distributed, for example, distributed around the user operation interface.
The voice processing module 13 is configured to recognize the voice instruction information collected by the voice collecting module 12 to obtain a voice recognition result.
And the data processing module 14 is configured to search for a first control command word matching the voice recognition result.
The control module 15 is configured to search for a first control function item corresponding to the first control command word in the data storage module 11 according to the first control command word, and control the intelligent household appliance to execute the first control function item.
According to the off-line voice control device for the intelligent household appliances, the control instruction packet comprising at least one control function item and at least two control instruction words corresponding to each control function item is locally stored, the voice instruction information of the recognition user is collected to obtain a voice recognition result, the control instruction words matched with the voice recognition result are searched, the control function items corresponding to the control instruction words are locally searched according to the control instruction words, the intelligent household appliances are controlled to execute the control function items, the control function items corresponding to the control instruction words are realized in an off-line mode, one control function item is controlled by the control instruction words, the effective rate of voice control is improved, and the control experience of the user on the intelligent household appliances is enhanced.
In an embodiment, the data storage module 11 is further configured to update the control instruction packet according to a control command word edited by a user and corresponding to a second control function item, where the second control function item is any control function item in the at least one control function item, that is, the second control function item may be the first control function item, and may also be another control function item.
The editing of the second control function item may be to modify, add, delete, and correspondingly store the edited control command word and the corresponding relationship between the control command word and the second control function item.
In one embodiment, as shown in fig. 2, the apparatus further comprises: and a display module 16, wherein the display module 16 may be a touch display screen (e.g., lcd touch display screen) disposed on the housing of the smart home device.
The display module 16 is configured to control a display page to display at least one control function item included in the control instruction packet and at least two control command words having a corresponding relationship with each control function item. Specifically, the display module 16 may display the at least one control function item and at least two control command words having a corresponding relationship with each control function item according to a user instruction. So that the user can know the corresponding relation between the control function item and the control command word.
For example, the display module 16 displays the correspondence between the voice command words stored in the data storage module 11 and the control function items matched with the voice command words on the screen, so that the user can edit, add and delete the voice command words on the screen, select the control function programs matched with the voice command words, test and store the custom modifications, and restore the factory settings of the voice control functions.
In an embodiment, the display module 16 is further configured to display a display page for editing the second control function item to the user, send a prompt for editing the second control function item, and acquire a control command word corresponding to the second control function item (any one of the control function items stored in the data storage module 11) edited by the user through the display page. Or, in this embodiment, when the display page displays at least one control function item included in the control instruction packet and at least two control command words having a corresponding relationship with each control function item, a prompt for editing the second control function item by voice is displayed to the user, and the voice acquisition module is further configured to receive the control command word corresponding to the second control function item (any one of the control function items stored in the data storage module 11) edited by voice by the user.
The editing of the second control function item may be to modify, add, delete, and correspondingly store the edited control command word and the corresponding relationship between the control command word and the second control function item.
In an embodiment, the data processing module 14 is further configured to obtain, through the voice collecting module 12, a control command word edited by the user in a voice manner, where the control command word corresponds to the second control function item (any one of the control function items stored in the data storage module 11). The editing of the second control function item may be to modify, add, delete, and correspondingly store the edited control command word and the corresponding relationship between the control command word and the second control function item.
In an embodiment, the data storage module 11 is further configured to, when the data processing module 11 does not find the first control command word matching the voice recognition result, extract the control command word from the voice recognition result, and update the control command packet according to a user instruction, where the extracted control command word corresponds to a third control function item in the at least one control function item.
Specifically, if the voice command sent by the user is matched with the corresponding control command word, a prompt is sent to the user to enable the voice command to correspond to the control function item, and the voice command and the control function item are corresponding and stored in the data storage module 11 according to the user operation. For example, if the user wants to increase the volume, the voice command is a loud sound, and the voice command is not matched with the corresponding increase volume control function, the voice command can be made loud and the increase volume can be corresponding according to the user operation.
In an embodiment, the data storage module 11 is further configured to obtain a command word grammar file for calibrating the command word combination rule, and determine a speech recognition model file related to the command word grammar file, where the speech recognition model file includes standard speech feature vectors of all control command words in the control instruction packet.
The present embodiment can use various command word grammar files, and here, an anbf grammar file is taken as an example.
The data storage module 11 locally stores files and software constants, for example, control command words preset by a home appliance manufacturer, command word grammar files for grammar construction at a voice recognition initialization stage, voice recognition model files for grammar construction, control programs associated with control functions of intelligent home appliance devices, and correspondence between control command words and control programs. The anbf grammar file marks the combination rule of the voice command words. The grammar construction is to analyze anbf grammar files by a compiling principle, and the voice recognition model files are a set of voice feature vectors of each voice command word in the grammar files.
In one embodiment, the voice processing module 13 is configured to convert voice instruction information into a voice feature vector, match the voice feature vector with a standard voice feature vector in the voice recognition model file, obtain a standard voice feature vector closest to the voice feature vector from the voice recognition model file, and determine the voice recognition result.
Specifically, the voice processing module 13 analyzes and processes the audio signal acquired by the voice acquisition module 12, obtains a group of voice feature vectors through encoding operation on the audio signal, compares and matches the voice feature vectors with model vectors in a voice recognition model file, obtains feature vectors closest to the voice feature vectors from the model file, and analyzes a voice recognition result through feature vector calculation.
In one embodiment, the speech recognition result obtained by the speech processing module 13 is a text;
the data processing module 14 is further configured to search the matching first control command word from the storage module according to the word if the word is included in the command word of the storage unit and the command word grammar file.
In one embodiment, the data storage module 11 is configured to assign an ID to each of the control function items; and establishing a storage table, wherein all control command words with the ID of each control function item and the corresponding relation of each control function item are correspondingly stored in the storage table in real time.
Specifically, taking a control function item as a software control program as an example, the data storage and storage module 11 defines an identifier id for each control program, for example, the volume-up id is 1, and the id corresponds to the volume-up control program; and the id for reducing the volume is 2, the id corresponds to a control program for reducing the volume, and when a control command word is newly added by a user to select a control function item, the id for actually selecting the control function item is matched and combined with the command word.
Therefore, the control function items can be conveniently corresponded to the control command words during storage, and the corresponding control function items can be conveniently called.
In one embodiment, the speech processing module 13 is further configured to determine a speech recognition threshold;
the data processing module is further configured to compare the voice recognition threshold with a preset threshold, where the voice recognition threshold is greater than the threshold, the voice recognition result is accurate, and the voice recognition threshold is less than the threshold, the voice recognition result is invalid.
Specifically, the voice processing module 13 obtains the voice recognition result and also obtains a voice recognition threshold, compares the threshold with a threshold preset by the system, where a threshold larger than the threshold indicates that the voice recognition result is accurate, and when a threshold smaller than the threshold indicates that the recognition result is invalid.
The off-line voice control device for the intelligent household appliances of the embodiment acquires voice instruction information and voice instruction information of a recognition user to obtain a voice recognition result by locally storing a control instruction packet containing at least one control function item and at least two control instruction words corresponding to each control function item, searches the control instruction words matched with the voice recognition result, locally searches the control function items corresponding to the control instruction words according to the control instruction words, and controls the intelligent household appliances to execute the control function items, so that the control function items corresponding to a plurality of control instruction words are realized in an off-line mode, one control function item is controlled by a plurality of control instruction words, the effective rate of voice control is improved, and the control experience of users on the intelligent household appliances is enhanced.
Example 2
In this embodiment, a data storage module 11, a voice processing module 13, a data processing module 14, and a control module 15 are disposed inside an intelligent household appliance, and a voice acquisition module 12 and a display module 16 are disposed on a housing of the intelligent household appliance or embedded in the housing.
The intelligent household appliance offline voice control device is applied to an offline mode (namely, a state of not being connected with a network) of an intelligent household appliance, and analyzes and processes collected voice signals locally without transmitting the voice signals to a cloud server for analysis and processing.
The data storage module 11 is used for local database storage, local file storage, and software constant storage, for example, control command words preset by a home appliance manufacturer, command word grammar files for grammar construction at a speech recognition initialization stage, speech recognition model files for grammar construction, control programs associated with intelligent home appliance control functions, and correspondence between control command words and control program items. The grammar file is an anbf grammar file, and the anbf grammar file marks the combination rule of the voice command words. The grammar construction analyzes the anbf grammar file through a compiling principle, and a voice recognition model file related to the grammar file is trained. The speech recognition model file is a set of speech feature vectors of all speech command words in the grammar file.
The data storage module 11 presets factory control command words and corresponding control function items of the intelligent household electrical appliance device, and anbf grammar files corresponding to the control command words. The data processing module 11 stores data for analyzing interaction between a user and the intelligent household appliance through software, and the user updates and modifies the data storage module 11 and the anbf grammar file.
The data storage module 11 presets default control command words and equipment control function programs of the system, stores the corresponding relation between the command words and the control function items in the data storage module 11, and the data storage module 11 stores anbf grammar files containing all default voice command words.
The control command words can be edited by a user, the control function program cannot be updated and modified by the user, and the anbf grammar file is synchronously updated and modified after the user updates and modifies the control command words.
The data storage module 11 is configured to assign an ID to each control function item; and establishing a storage table, wherein the ID of each control function item is stored in the storage table in real time. Specifically, taking a control function item as a software control program as an example, the data storage and storage module 11 defines an identifier id for each control program, for example, the volume-up id is 1, and the id corresponds to the volume-up control program; and the id for reducing the volume is 2, the id corresponds to a control program for reducing the volume, and when a control command word is newly added by a user to select a control function item, the id for selecting the control function item is matched and combined with the voice command word.
The data storage module 11 reconstructs an anbf grammar file, a grammar compiling engine compiles the anbf grammar file to generate a voice recognition model file, standard voice feature vectors of all voice command words exist in the model file, and the voice feature vectors of the command words output by the voice processing single module 13 are sequentially compared with the feature vectors in the model file to obtain the feature vector with the highest matching degree.
The display module 16 may be an lcd touch screen for interaction between a user and stored data of the device, the corresponding relationship between the control command word and the control function item (control program) is displayed to the user through the display module 16, the user may modify, newly add, delete and store the control command word through the display module 16, select the control function item (control program) associated with the control command word, test and store the custom modification, and restore factory settings of the voice control function.
The display module 16 receives control command words for modification, addition and deletion of the user on the editing page, wherein the length of the modified and added command words can be 4-6 Chinese characters. In this embodiment, in order to increase the recognition degree, the voice command words are modified to avoid the same or similar pronunciation, and a prompt may be issued if the same or similar pronunciation is used. After a voice command word is newly added, determining a control function item corresponding to the voice command word according to user selection, wherein one voice command word only corresponds to one control function, and one control function can be corresponding to a plurality of voice command words.
The voice acquisition module 11 is a dual-microphone voice acquisition device, and is configured to acquire voice instruction audio information of a user, and transmit the acquired audio information to the voice processing module 13. Specifically, the voice processing module 13 analyzes and processes the audio signal acquired by the voice acquisition module 12, obtains a group of voice feature vectors through encoding operation on the audio signal, compares and matches the voice feature vectors with model vectors in a voice recognition model file, obtains feature vectors closest to the voice feature vectors from the model file, and analyzes a voice recognition result through feature vector calculation.
The voice processing module 13 analyzes the audio signal collected by the microphone through software, encodes the audio signal into a group of voice feature vectors, and analyzes and calculates the voice recognition result by comparing the analyzed voice feature vectors with the feature vectors in the model file.
In this embodiment, the voice recognition result may be a text word, the text word is included in the command word of the data storage module 11 and the anbf grammar file, and the id of the control function is matched from the data storage module 11 according to the text word, that is, the id is transmitted to the control module 15 to realize accurate control of the device.
In this embodiment, the voice processing module 13 obtains the voice recognition result and also obtains a voice recognition threshold, compares the threshold with a threshold preset by the system, where a threshold larger than the threshold indicates that the voice recognition result is accurate, and when a threshold smaller than the threshold indicates that the recognition result is invalid.
The data processing module 14 is used for realizing the operations of updating, newly adding, deleting, saving and the like of the data storage module 11, updating and modifying the anbf grammar file, reconstructing the updated anbf grammar file to generate a voice recognition model file, and analyzing and processing a voice recognition result.
The control module 15 controls the intelligent household electrical appliance to execute corresponding control operation items according to the device control program preset in the data storage module 11.
The use scenario of this embodiment is that the volume of a loudspeaker of an intelligent household appliance (such as a refrigerator) is adjusted, control command words preset by a manufacturer are volume increase and volume decrease, the preset control command words are pre-stored in the data storage module 11 and written in an anbf grammar file, the preset control command words are associated with a function program for adjusting the volume in the data storage module 11, and when the grammar is constructed in the speech recognition initialization stage, a generated speech recognition model file includes speech feature vectors of the volume increase and volume decrease command words. When the user uses the voice pitch to save the voice volume, the user speaks the voice command of increasing the voice volume and decreasing the voice volume to be recognized and executed. When a user or a family member feels that the user or the family member increases the volume and reduces the volume and speaks more diversely and is not suitable for the usual speaking habit, the data storage module 11 can update the anbf grammar file content while the display unit newly edits or adds voice command words suitable for the speaking habit, such as one louder and one louder, and edits or newly adds command words. After adding a new command word, the user selects a control program matched with the new voice command word in the function list, the data storage module 11 stores the matching relationship between the voice command word and the control program, the data storage module 11 reconstructs the anbf grammar to generate a new voice recognition model file, the previous voice recognition model file is covered, and the voice characteristics of the command word with a loud voice and a small voice are constructed into the new voice recognition model file. After the speech recognition is reinitialized, the user speaks a little loud, a little soft and other command words to be recognized by the speech recognition module 13, so that the speech recognition result is analyzed by the computing unit, and the volume of the loudspeaker of the refrigerator can be controlled by combining the control module 15.
According to the off-line voice control device for the intelligent household appliance, voice command words before modification and deletion of a user cannot be constructed into a voice recognition model file in a grammar construction stage, a recognition result cannot be obtained in voice feature matching and analysis calculation, the voice command words after modification and newly added by the user are constructed into the voice recognition model file in the grammar construction stage, the recognition result is easily obtained in the voice feature matching and analysis calculation, and the technical problem that the user cannot timely speak a control instruction or speak an incorrect control instruction due to the fact that the control command words preset by the intelligent household appliance cannot be remembered or do not conform to speaking habits of the user in the process of controlling the intelligent household appliance by using the voice command can be solved.
The off-line voice control device for the intelligent household appliances of the embodiment acquires voice instruction information and voice instruction information of a recognition user to obtain a voice recognition result by locally storing a control instruction packet containing a plurality of control function items and at least two control instruction words corresponding to each control function item, searches the control instruction words matched with the voice recognition result, locally searches the control function items corresponding to the control instruction words according to the control instruction words, and controls the intelligent household appliances to execute the control function items, so that the control function items corresponding to the control instruction words are realized in an off-line mode, one control function item is controlled by the control instruction words, the effective rate of voice control is improved, and the control experience of the user on the intelligent household appliances is enhanced.
Example 3
This embodiment provides a method for performing offline voice control on an intelligent home appliance by using the offline voice control device for an intelligent home appliance provided in the foregoing embodiment 1 and embodiment 2, and as shown in fig. 4, the offline voice control method for an intelligent home appliance includes:
401. and displaying a voice command word editing interface according to the user instruction, wherein the editing interface displays the voice command word and the equipment control function corresponding to the voice command word.
402. And detecting a voice command word editing instruction of the user, and modifying, adding or deleting the voice command word.
403a, modifying the voice command word; 403b, adding a voice command word; 403c, deleting the voice command word.
And 403d, selecting the control function item matched with the newly added command word.
This step performs matching according to user instructions.
404. And updating the corresponding relation between the voice command words and the control function items.
This step is implemented by the data storage module 11.
405. And updating the anbf grammar file.
The specific steps are reading the anbf grammar file content, and updating and saving according to the existing voice command words of the data storage module 11.
406. And reconstructing the anbf grammar file, generating a new voice recognition resource file, and covering the previously generated voice recognition model file.
This step is implemented by the data storage module 11.
407. And detecting a user voice control instruction.
This step is implemented by the voice lake zone module 12.
408. And analyzing and calculating the voice characteristic information of the voice of the user.
This step is implemented by the speech processing module 13.
409. And sequentially comparing the acquired voice characteristic information with each voice characteristic in the voice recognition model file, and acquiring the voice characteristic with the highest matching degree from the voice recognition model file to obtain a voice recognition result.
This step is implemented by the speech processing module 13.
410. And comparing the threshold value of the voice recognition result with a preset voice recognition threshold value.
411. The comparison is made if the threshold value > the threshold value.
If the threshold > the threshold value 412 is executed, otherwise, the procedure is ended, which is implemented by the speech processing module 13.
412. And matching a control function item according to the voice recognition result, and controlling the equipment according to the control function item.
Specifically, the control function item is matched from the data storage module 11 according to the voice recognition result, and the control of the equipment is realized according to the control function item
This step is realized by the data processing module 14 and the control module 15.
The application scenario of this embodiment is that when a user experiences that some voice command words preset by a home appliance manufacturer or modified before do not conform to the usual speaking habits of the user during the process of using the voice to control the intelligent home appliance, a command word editing page can be entered on the lcd display screen, the editing page displays all voice command words of the local data storage module, and matches device control function items corresponding to each voice command word, for example, a volume increase command word corresponds to a function of controlling the sound increase of a speaker. The user can modify, add, delete and other modification operations on the command word on the editing page, and the user modifies the voice command word without changing the control function corresponding to the command word; a user adds voice command words in the data storage module 11 and selects control function items matched with the voice command words, wherein one voice command word only corresponds to one control function item, and one control function item can correspond to a plurality of different voice command words; the user deletes the voice command word from the data storage module 11 and releases the control function item corresponding thereto. The user modified and newly added command words take 4-6 Chinese characters as the best recognition results, and the voice command words with similar pronunciation can be avoided in order to increase the recognition success rate. And after finishing the page modification operation, updating the voice command words of the data storage and storage module and updating the corresponding relation between the command words and the control function items. And updating the local anbf grammar file according to the voice command words of the data storage module 11, after the anbf grammar file is updated and stored, reconstructing the grammar of the anbf grammar file, and regenerating the model file related to the voice recognition. When a user speaks a voice command word, the voice acquisition module 12 acquires voice information of the user and transmits the voice information to the voice processing module 13, the voice processing module 13 performs noise reduction and echo cancellation on the received audio signal and then performs coding operation on the audio signal to obtain a group of feature vectors, namely voice feature vectors, the voice feature vectors are compared and matched with the voice feature vectors in the voice recognition model file, the feature vectors closest to the voice feature vectors are obtained from the model file, and software calculates and analyzes a voice recognition result through the feature vectors. The speech processing module 13 will attach a matching score of the diagnosis-specific vector, i.e. a threshold value, while outputting the speech recognition result, where the threshold value is used to characterize the matching degree between the speech recognition result and the speech, and the larger the threshold value is, the more accurate the speech recognition result is. And presetting a threshold (which can be changed), wherein the voice recognition result is accurate when the threshold is larger than the threshold, and the recognition result is invalid when the threshold is smaller than the threshold. And when the threshold value is larger than the threshold value, matching a control function from the data storage module 11 according to the voice recognition result, and sending the control function to the control module 15 to realize the control of the intelligent household appliance.
The intelligent household appliance offline voice control method of the embodiment is applied to the intelligent household appliance offline voice control device of the embodiment, so that control function items corresponding to a plurality of control command words are realized in an offline mode, and one control function item is controlled by a plurality of control command words, so that the voice control efficiency is improved, and the control experience of a user on the intelligent household appliance is enhanced.
Having described embodiments of the present invention, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. The utility model provides an intelligence household electrical appliances off-line voice control device, its characterized in that, the device is used for intelligence household electrical appliances off-line voice control, and the device includes: the voice recognition system comprises a data storage module, a voice acquisition module, a voice processing module, a data processing module and a control module;
the data storage module is arranged locally and used for storing a control instruction packet locally, and the control instruction packet comprises at least one control function item and at least two control command words which correspond to the control function items;
the voice acquisition module is used for acquiring voice instruction information of a user when the intelligent household appliance enters a voice acquisition mode and transmitting the voice instruction information to the voice processing module;
the voice processing module is used for recognizing the voice instruction information to obtain a voice recognition result;
the data processing module is used for searching a first control command word matched with the voice recognition result;
the control module is used for searching a first control function item corresponding to the first control command word in the data storage module according to the first control command word, and controlling the intelligent household appliance to execute the first control function item.
2. The apparatus according to claim 1, wherein the data storage module is further configured to update the control instruction packet according to a control command word edited by a user and corresponding to a second control function item, where the second control function item is any control function item in the at least one control function item.
3. The apparatus of claim 2, further comprising: a display module;
the display module is used for controlling a display page to display at least one control function item contained in the control instruction packet and at least two control command words which have corresponding relations with each control function item;
the display module is further used for displaying a display page for editing the second control function item to a user, sending a prompt for editing the second control function item, and acquiring a control command word corresponding to the second control function item edited by the user through the display page; or when the display page displays at least one control function item contained in the control instruction packet and at least two control command words having a corresponding relation with each control function item, a prompt of the second control function item through voice editing is displayed for a user, and the voice acquisition module is further used for receiving the control command words corresponding to the second control function item through voice editing by the user.
4. The apparatus according to claim 2, wherein the data processing module is further configured to obtain, through the voice capture module, a control command word corresponding to the second control function item edited by the user through voice.
5. The apparatus according to claim 2, wherein the data storage module is further configured to, when the data processing module does not find the first control command word matching the voice recognition result, extract a control command word from the voice recognition result, and update the control command packet according to a user instruction by corresponding the extracted control command word to a third control function item in the at least one control function item.
6. The apparatus of claim 1, wherein the data storage module is further configured to obtain a command word grammar file for calibrating the command word combination rule, and determine a speech recognition model file associated with the command word grammar file, where the speech recognition model file contains standard speech feature vectors of all control command words in the control instruction packet.
7. The apparatus of claim 6, wherein the speech processing module is configured to convert speech instruction information into speech feature vectors, match the speech feature vectors with standard speech feature vectors in the speech recognition model file, obtain standard speech feature vectors closest to the speech feature vectors from the speech recognition model file, and determine the speech recognition result.
8. The apparatus according to claim 6 or 7, wherein the speech recognition result obtained by the speech processing module is a text;
and the data processing module is also used for searching the matched first control command word from the storage module according to the characters if the characters are contained in the command words of the storage unit and the command word grammar file.
9. The apparatus according to any one of claims 1 to 7, wherein the data storage module is configured to assign an ID to each of the control function items; and establishing a storage table, wherein all control command words with the ID of each control function item and the corresponding relation of each control function item are correspondingly stored in the storage table in real time.
10. The apparatus of any of claims 1-7, wherein the speech processing module is further configured to determine a speech recognition threshold;
the voice processing module is further configured to compare the voice recognition threshold with a preset threshold, where the voice recognition threshold is greater than the threshold, the voice recognition result is accurate, and the voice recognition threshold is less than the threshold, the voice recognition result is invalid.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911395573.8A CN111161731A (en) | 2019-12-30 | 2019-12-30 | Intelligent off-line voice control device for household electrical appliances |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911395573.8A CN111161731A (en) | 2019-12-30 | 2019-12-30 | Intelligent off-line voice control device for household electrical appliances |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111161731A true CN111161731A (en) | 2020-05-15 |
Family
ID=70559020
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911395573.8A Pending CN111161731A (en) | 2019-12-30 | 2019-12-30 | Intelligent off-line voice control device for household electrical appliances |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111161731A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112581952A (en) * | 2020-11-09 | 2021-03-30 | 金茂智慧科技(广州)有限公司 | Intelligent equipment control method combining online voice with offline voice |
| CN112885347A (en) * | 2021-01-22 | 2021-06-01 | 海信电子科技(武汉)有限公司 | Voice control method of display device, display device and server |
| CN115064171A (en) * | 2022-08-18 | 2022-09-16 | 安徽立诺威智能科技有限公司 | Voice awakening method and system for intelligent air disinfection equipment |
| CN117784632A (en) * | 2024-02-28 | 2024-03-29 | 深圳市轻生活科技有限公司 | Intelligent household control system based on offline voice recognition |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104978957A (en) * | 2014-04-14 | 2015-10-14 | 美的集团股份有限公司 | Voice control method and system based on voiceprint identification |
| CN105096940A (en) * | 2015-06-30 | 2015-11-25 | 百度在线网络技术（北京）有限公司 | Method and device for voice recognition |
| CN106550132A (en) * | 2016-10-25 | 2017-03-29 | 努比亚技术有限公司 | A kind of mobile terminal and its control method |
| CN106817144A (en) * | 2017-01-18 | 2017-06-09 | 上海洪洋通信科技有限公司 | Various dimensions robot system |
| CN108364644A (en) * | 2018-01-17 | 2018-08-03 | 深圳市金立通信设备有限公司 | A kind of voice interactive method, terminal and computer-readable medium |
| CN108731183A (en) * | 2018-05-31 | 2018-11-02 | 奥克斯空调股份有限公司 | A kind of intelligent air-conditioning system and air-conditioning |
| CN108766441A (en) * | 2018-05-29 | 2018-11-06 | 广东声将军科技有限公司 | A kind of sound control method and device based on offline Application on Voiceprint Recognition and speech recognition |
| CN109036393A (en) * | 2018-06-19 | 2018-12-18 | 广东美的厨房电器制造有限公司 | Wake-up word training method, device and the household appliance of household appliance |
| CN109495360A (en) * | 2018-12-18 | 2019-03-19 | 深圳国美云智科技有限公司 | A kind of smart home Internet of Things platform, offline sound control method and system |
-
2019
- 2019-12-30 CN CN201911395573.8A patent/CN111161731A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104978957A (en) * | 2014-04-14 | 2015-10-14 | 美的集团股份有限公司 | Voice control method and system based on voiceprint identification |
| CN105096940A (en) * | 2015-06-30 | 2015-11-25 | 百度在线网络技术（北京）有限公司 | Method and device for voice recognition |
| CN106550132A (en) * | 2016-10-25 | 2017-03-29 | 努比亚技术有限公司 | A kind of mobile terminal and its control method |
| CN106817144A (en) * | 2017-01-18 | 2017-06-09 | 上海洪洋通信科技有限公司 | Various dimensions robot system |
| CN108364644A (en) * | 2018-01-17 | 2018-08-03 | 深圳市金立通信设备有限公司 | A kind of voice interactive method, terminal and computer-readable medium |
| CN108766441A (en) * | 2018-05-29 | 2018-11-06 | 广东声将军科技有限公司 | A kind of sound control method and device based on offline Application on Voiceprint Recognition and speech recognition |
| CN108731183A (en) * | 2018-05-31 | 2018-11-02 | 奥克斯空调股份有限公司 | A kind of intelligent air-conditioning system and air-conditioning |
| CN109036393A (en) * | 2018-06-19 | 2018-12-18 | 广东美的厨房电器制造有限公司 | Wake-up word training method, device and the household appliance of household appliance |
| CN109495360A (en) * | 2018-12-18 | 2019-03-19 | 深圳国美云智科技有限公司 | A kind of smart home Internet of Things platform, offline sound control method and system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112581952A (en) * | 2020-11-09 | 2021-03-30 | 金茂智慧科技(广州)有限公司 | Intelligent equipment control method combining online voice with offline voice |
| CN112885347A (en) * | 2021-01-22 | 2021-06-01 | 海信电子科技(武汉)有限公司 | Voice control method of display device, display device and server |
| CN115064171A (en) * | 2022-08-18 | 2022-09-16 | 安徽立诺威智能科技有限公司 | Voice awakening method and system for intelligent air disinfection equipment |
| CN117784632A (en) * | 2024-02-28 | 2024-03-29 | 深圳市轻生活科技有限公司 | Intelligent household control system based on offline voice recognition |
| CN117784632B (en) * | 2024-02-28 | 2024-05-14 | 深圳市轻生活科技有限公司 | Intelligent household control system based on offline voice recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111161731A (en) | Intelligent off-line voice control device for household electrical appliances | |
| CN107454508B (en) | TV set and TV system of microphone array | |
| CN108831469B (en) | Voice command customizing method, device and equipment and computer storage medium | |
| JP6482911B2 (en) | Device control method and electric device | |
| CN107644638B (en) | Audio recognition method, device, terminal and computer readable storage medium | |
| KR102004872B1 (en) | Electronic device, server and control methods thereof | |
| CN109378006B (en) | Cross-device voiceprint recognition method and system | |
| JP2014089437A (en) | Voice recognition device, and voice recognition method | |
| KR102411619B1 (en) | Electronic apparatus and the controlling method thereof | |
| CN105791931A (en) | Smart television and voice control method of the smart television | |
| CN107958668B (en) | Voice control broadcasting method and voice control broadcasting system of smart television | |
| CN104978964B (en) | Phonetic control command error correction method and system | |
| JP7159475B2 (en) | Voice control method, cloud server and terminal device | |
| KR20140089871A (en) | Interactive server, control method thereof and interactive system | |
| JP2014132370A (en) | Image processing apparatus, control method thereof, and image processing system | |
| KR20140092634A (en) | Electronic apparatus and method of controlling the same | |
| KR20150089145A (en) | display apparatus for performing a voice control and method therefor | |
| US20160210966A1 (en) | Voice recognition processing device, voice recognition processing method, and display device | |
| JP6675078B2 (en) | Misrecognition and correction method, misrecognition and correction device, and misrecognition and correction program | |
| KR102501655B1 (en) | Display apparatus, voice acquiring apparatus and voice recognition method thereof | |
| KR20170051994A (en) | speech recognition device and here of control Method | |
| EP2757557B1 (en) | Electronic apparatus and voice processing method thereof | |
| EP4184309A1 (en) | Display device | |
| CN109739462A (en) | A kind of method and device of content input | |
| US20180332339A1 (en) | Set-Top Box with Enhanced Functionality and System and Method for Use of Same |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200515 |