CN111415657A - Decentralized device, multi-device system and voice control method thereof - Google Patents
Decentralized device, multi-device system and voice control method thereof Download PDFInfo
- Publication number
- CN111415657A CN111415657A CN201910011444.8A CN201910011444A CN111415657A CN 111415657 A CN111415657 A CN 111415657A CN 201910011444 A CN201910011444 A CN 201910011444A CN 111415657 A CN111415657 A CN 111415657A
- Authority
- CN
- China
- Prior art keywords
- energy value
- module
- devices
- voice
- awakening
- 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
- 238000000034 method Methods 0.000 title claims description 20
- 238000004364 calculation method Methods 0.000 claims abstract description 23
- 238000012216 screening Methods 0.000 claims abstract description 16
- 238000004891 communication Methods 0.000 claims abstract description 5
- 230000005236 sound signal Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 abstract description 9
- 238000010586 diagram Methods 0.000 description 2
- 210000004072 lung Anatomy 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000007306 turnover Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000002618 waking 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
- G10L15/28—Constructional details of speech recognition systems
- G10L15/32—Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems
-
- 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
A decentralized device comprises a voice recognition module, an energy value calculation module, an energy value screening and sorting module, a transceiver module and an execution module, wherein the energy value calculation module, the transceiver module and the execution module are in signal connection with the energy value screening and sorting module, the energy value calculation module is also in signal connection with the voice recognition module and the transceiver module, and the transceiver modules of different devices can be in mutual communication. A decentralized multi-device system comprising a plurality of devices as described above. According to the invention, the multi-device identification adopts a decentralized processing mode, so that centralized processing equipment is removed, the identification is converted into self-processing of the equipment, and complex calculation is decomposed into a plurality of devices to be carried out in parallel, the calculation speed is improved, the network bandwidth requirement is reduced, the response speed of the system is improved, and the experience of a user is improved.
Description
Technical Field
The invention belongs to the field of artificial intelligence, relates to a voice recognition technology, and particularly relates to a decentralized device, a multi-device system and a voice control method thereof.
Background
With the advent of speech recognition functionality, there may be many intelligent speech devices in the same scene, such as a lobby, that need to be awakened with an awakening word when in use. However, if the wake-up words of these devices are the same, then it may happen that multiple devices respond to the wake-up word of the user, interfering with the real needs of the user. For example, in one particular scenario shown in fig. 1, a user utters a voice command, and the wake-up words of devices a1, a2, A3 are identical, and the command may be recognized and executed.
In order to solve the above problems, a common method in the prior art is to uniformly determine which device a user is close to by a processing center, such as a cloud server, and which device responds to a wake-up instruction of the user, but this method needs to use an additional server for determination, which increases the cost; in the prior art, calculation and sequencing can be performed by directly receiving voice information and then calculating, and a device for executing a command is selected by calculating the distance between each device.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention discloses a decentralized device, a multi-device system and a voice control method thereof.
The decentralized equipment comprises a voice recognition module, an energy value calculation module, an energy value screening and sorting module, a transceiver module and an execution module, wherein the energy value calculation module, the transceiver module and the execution module are in signal connection with the energy value screening and sorting module, the energy value calculation module is also in signal connection with the voice recognition module and the transceiver module, and the transceiver modules of different equipment can be in mutual communication.
The decentralized multi-device system comprises a plurality of devices, wherein the awakening words of at least two devices in the plurality of devices are the same.
The invention also discloses a decentralized multi-equipment system voice control method, which comprises the following steps:
s1 the device recognizes the awakening word WS in the voice;
s2, the device calculates the energy value PA of the voice signal received by itself, and packs the calculation result and the identified awakening word WS to send to all devices in the system, and meanwhile receives the data packet containing the awakening word and the energy value sent by other devices in the system;
s3, after the awakening words in all the data packets are identified, the awakening words matched with the equipment are screened out;
the screening range comprises a self-identified awakening word WS and a received awakening word WS;
s4, sorting the screened awakening words according to the energy value, executing the command if the voice signal energy value PA corresponding to the awakening word identified by the equipment with the highest ranking is obtained, and otherwise, terminating the process.
Preferably, the energy value is an audio signal energy value or a signal-to-noise ratio of the audio signal energy to the environmental noise.
Preferably, after the device identifies the wake word WS in the voice in S1, the device first compares whether the wake word WS matches itself, if yes, the subsequent steps are continued, otherwise, the subsequent steps are terminated.
According to the invention, the multi-device identification adopts a decentralized processing mode, so that centralized processing equipment is removed, the identification is converted into self-processing of the equipment, and complex calculation is decomposed into a plurality of devices to be carried out in parallel, the calculation speed is improved, the network bandwidth requirement is reduced, the response speed of the system is improved, and the experience of a user is improved.
Drawings
FIG. 1 is a schematic diagram of an exemplary embodiment of the present invention; the letters a1, a2, A3, B in fig. 1 denote different devices.
Fig. 2 shows a schematic diagram of the internal modules of an embodiment of the apparatus according to the present invention.
Fig. 3 is a flow chart illustrating a speech control method according to an embodiment of the present invention.
Detailed Description
The following provides a more detailed description of the present invention.
The decentralized device, the multi-device system and the voice control method thereof are mainly used for solving the problem of selecting the device for executing the command from the devices with the same awakening words selectively.
For example, in the scenario shown in fig. 1, devices a1, a2, A3, and B exist, where the wakeup words of devices a1, a2, and A3 are all a0, and the wakeup word of device B is B0, and after the user utters the wakeup word a0, all devices a1, a2, and A3 recognize the wakeup word, but generally, the true purpose of the user is to want the device a1 closest to the user to act.
The wake-up word is a special word for waking up the device in voice recognition, each device has a unique wake-up word, and in a voice recognition environment, the device firstly recognizes the wake-up word and verifies that the wake-up word is matched with the device, and then the device can continue to perform subsequent command recognition or command execution and other operations.
In order to solve the problems, the invention discloses decentralized equipment which comprises a voice recognition module, an energy value calculation module, an energy value screening and sorting module, a transmitting and receiving module and an execution module, wherein the energy value calculation module, the transmitting and receiving module and the execution module are all in signal connection with the energy value screening and sorting module, the energy value calculation module is also in signal connection with the voice recognition module and the transmitting and receiving module, and the transmitting and receiving modules of different equipment can be communicated with each other.
The system is applied to a system which is composed of a plurality of devices and has at least two devices with the same awakening words, such as a plurality of ceiling lamps with different positions in an auditorium, air conditioners in different rooms in the same house and the like.
The voice control method of the decentralized multi-device system comprises the following steps
S1 the device recognizes the awakening word WS in the voice;
s2, the device calculates the energy value PA of the voice signal received by itself, and packs the calculation result and the identified awakening word WS to send to all devices in the system, and meanwhile receives the data packet containing the awakening word and the energy value sent by other devices in the system;
s3, after the awakening words in all the data packets are identified, the awakening words matched with the equipment are screened out;
the screening range comprises a self-identified awakening word WS and a received awakening word WS;
s4, sorting the screened awakening words according to the energy value, executing the command if the voice signal energy value PA corresponding to the awakening word identified by the equipment with the highest ranking is obtained, and otherwise, terminating the process.
The speech recognition process applying the present invention is illustrated by taking the application scenario of fig. 1 as an example as follows:
a user sends out a wake-up word A0, after the wake-up word is recognized by each device A1, A2, A3 and B in the system, wherein after the A1, A2 and A3 receive the wake-up word, the energy values of the voice signals received by the devices are calculated to be PA1, PA2 and PA3 respectively, and the recognized wake-up word and the energy values are packaged and sent through a transceiving module respectively.
The energy value is typically an audio signal energy value or a signal-to-noise ratio of the audio signal energy to the ambient noise, the energy value being proportional to the square of the amplitude of the waveform.
The transceiver module directly transmits the information after identification to the transceiver module of other equipment, the information is transmitted only by few bytes generally, an indoor short-distance wireless transmission mode such as Bluetooth and the like is adopted, and generally, no extra wire or larger bandwidth is adopted.
The communication between the transceiver modules of each device can be carried out in a wired or wireless mode, and for wireless short-distance application, the communication between the transceiver modules can be carried out in modes such as Bluetooth, WIFI, 2.4G wireless modules and the like.
The voice recognition method and the voice recognition device are based on the premise that the voice device is locally recognized, voice information does not need to be uploaded to a cloud end or other central servers for recognition and turnover, the voice recognition module of the device can utilize a CI1006 series voice recognition chip of the applicant's Qingtai Lung company to perform local recognition, and functions of energy value calculation, energy value screening and sorting and the like can also be realized in the CI1006 series voice recognition chip of the applicant's QingTai Lung company by utilizing writing software.
For the device B of which the self-awakening word is not A0, the device B can be calculated and sent as A1, A2 and A3, and whether the device B is matched with the device B or not can be firstly compared, if yes, the subsequent steps are continued, and if not, the subsequent steps are terminated, so that unnecessary power consumption is saved.
After the device a1 calculates the self-received wake word energy value PA1, it sorts the received wake word energy values PA2 and PA3 transmitted by the devices a2 and A3 in the energy value screening and sorting module, if the PA1 energy value is found to be the highest, the device a1 executes the command, otherwise, the process is not executed and terminated.
Among the devices a1, a2, A3, the device whose energy value recognized by the self is the highest executes the command.
For the condition of simultaneously receiving a plurality of different awakening words, after receiving various awakening words, the energy value screening and sorting module firstly screens the awakening words matched with the equipment, wherein the screening range comprises the awakening words WS identified by the energy value screening and sorting module and then sorts the awakening words.
For a user, when sending a command, the user generally wants to control a device closest to the user, the closer the device is, the less the audio signal is attenuated on a propagation path, the higher the received energy value is, and the device needing to execute the command is defined through energy value sequencing.
According to the invention, the multi-device identification adopts a decentralized processing mode, so that centralized processing equipment is removed, the identification is converted into self-processing of the equipment, and complex calculation is decomposed into a plurality of devices to be carried out in parallel, the calculation speed is improved, the network bandwidth requirement is reduced, the response speed of the system is improved, and the experience of a user is improved.
The foregoing is a description of preferred embodiments of the present invention, and the preferred embodiments in the preferred embodiments may be combined and combined in any combination, if not obviously contradictory or prerequisite to a certain preferred embodiment, and the specific parameters in the examples and the embodiments are only for the purpose of clearly illustrating the inventor's invention verification process and are not intended to limit the patent protection scope of the present invention, which is defined by the claims and the equivalent structural changes made by the content of the description of the present invention are also included in the protection scope of the present invention.
Claims (5)
1. A decentralized device is characterized by comprising a voice recognition module, an energy value calculation module, an energy value screening and sorting module, a transceiver module and an execution module, wherein the energy value calculation module, the transceiver module and the execution module are in signal connection with the energy value screening and sorting module, the energy value calculation module is also in signal connection with the voice recognition module and the transceiver module, and the transceiver modules of different devices can be in mutual communication.
2. A decentralized multi-device system comprising a plurality of devices according to claim 1, wherein the wake-up words of at least two of said devices are the same.
3. A voice control method for a decentralized multi-device system is characterized by comprising the following steps:
s1 the device recognizes the awakening word WS in the voice;
s2, the device calculates the energy value PA of the voice signal received by itself, and packs the calculation result and the identified awakening word WS to send to all devices in the system, and meanwhile receives the data packet containing the awakening word and the energy value sent by other devices in the system;
s3, after the awakening words in all the data packets are identified, the awakening words matched with the equipment are screened out;
the screening range comprises a self-identified awakening word WS and a received awakening word WS;
s4, sorting the screened awakening words according to the energy value, executing the command if the voice signal energy value PA corresponding to the awakening word identified by the equipment with the highest ranking is obtained, and otherwise, terminating the process.
4. The decentralized multi-device speech control method according to claim 3, wherein the energy value is an audio signal energy value or a signal-to-noise ratio of audio signal energy to ambient noise.
5. The decentralized multi-device voice control method according to claim 3, wherein the device in S1 recognizes the wake word WS in the voice and then first compares whether it matches itself, if so, continues the subsequent steps, otherwise, terminates the subsequent steps.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910011444.8A CN111415657A (en) | 2019-01-07 | 2019-01-07 | Decentralized device, multi-device system and voice control method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910011444.8A CN111415657A (en) | 2019-01-07 | 2019-01-07 | Decentralized device, multi-device system and voice control method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111415657A true CN111415657A (en) | 2020-07-14 |
Family
ID=71493995
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910011444.8A Pending CN111415657A (en) | 2019-01-07 | 2019-01-07 | Decentralized device, multi-device system and voice control method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111415657A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112652310A (en) * | 2020-12-31 | 2021-04-13 | 乐鑫信息科技（上海）股份有限公司 | Distributed speech processing system and method |
| CN113010216A (en) * | 2021-03-12 | 2021-06-22 | 深圳市超维实业有限公司 | Ceiling lamp based awakening event management method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100004930A1 (en) * | 2008-07-02 | 2010-01-07 | Brian Strope | Speech Recognition with Parallel Recognition Tasks |
| CN106030699A (en) * | 2014-10-09 | 2016-10-12 | 谷歌公司 | Hotword detection on multiple devices |
| US20170083285A1 (en) * | 2015-09-21 | 2017-03-23 | Amazon Technologies, Inc. | Device selection for providing a response |
| CN107919119A (en) * | 2017-11-16 | 2018-04-17 | 百度在线网络技术（北京）有限公司 | Method, apparatus, equipment and the computer-readable medium of more equipment interaction collaborations |
| CN108766432A (en) * | 2018-07-02 | 2018-11-06 | 珠海格力电器股份有限公司 | Method for controlling cooperative work among household appliances |
| CN108766422A (en) * | 2018-04-02 | 2018-11-06 | 青岛海尔科技有限公司 | Response method, device, storage medium and the computer equipment of speech ciphering equipment |
-
2019
- 2019-01-07 CN CN201910011444.8A patent/CN111415657A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100004930A1 (en) * | 2008-07-02 | 2010-01-07 | Brian Strope | Speech Recognition with Parallel Recognition Tasks |
| CN106030699A (en) * | 2014-10-09 | 2016-10-12 | 谷歌公司 | Hotword detection on multiple devices |
| US20170083285A1 (en) * | 2015-09-21 | 2017-03-23 | Amazon Technologies, Inc. | Device selection for providing a response |
| CN107919119A (en) * | 2017-11-16 | 2018-04-17 | 百度在线网络技术（北京）有限公司 | Method, apparatus, equipment and the computer-readable medium of more equipment interaction collaborations |
| CN108766422A (en) * | 2018-04-02 | 2018-11-06 | 青岛海尔科技有限公司 | Response method, device, storage medium and the computer equipment of speech ciphering equipment |
| CN108766432A (en) * | 2018-07-02 | 2018-11-06 | 珠海格力电器股份有限公司 | Method for controlling cooperative work among household appliances |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112652310A (en) * | 2020-12-31 | 2021-04-13 | 乐鑫信息科技（上海）股份有限公司 | Distributed speech processing system and method |
| CN112652310B (en) * | 2020-12-31 | 2024-08-09 | 乐鑫信息科技（上海）股份有限公司 | Distributed speech processing system and method |
| CN113010216A (en) * | 2021-03-12 | 2021-06-22 | 深圳市超维实业有限公司 | Ceiling lamp based awakening event management method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111223497B (en) | Nearby wake-up method and device for terminal, computing equipment and storage medium | |
| CN106910500B (en) | Method and device for voice control of device with microphone array | |
| US11354089B2 (en) | System and method for dialog interaction in distributed automation systems | |
| CN111566730B (en) | Voice command processing in low power devices | |
| CN201129826Y (en) | Air conditioner control device | |
| CN107767867B (en) | Method, device, system and storage medium for realizing voice control network | |
| CN108320749A (en) | Far field voice control device and far field speech control system | |
| CN108592349A (en) | A kind of air-conditioner control system | |
| CN109949801A (en) | A kind of smart home device sound control method and system based on earphone | |
| CN111415657A (en) | Decentralized device, multi-device system and voice control method thereof | |
| CN110782897B (en) | Voice terminal communication method and system based on natural semantic coding | |
| CN118053426A (en) | Interconnection mutual control intelligent wireless switch and off-line voice control system thereof | |
| CN112585675A (en) | Method, apparatus and system for selectively using a plurality of voice data receiving apparatuses for intelligent service | |
| CN107529146B (en) | Multi-sensing indoor positioning method, device and system combined with audio and storage medium | |
| WO2024002298A1 (en) | Voice instruction processing method, apparatus and system, and storage medium | |
| KR20050030906A (en) | System for collectively recognizing speech and method thereof | |
| CN206322472U (en) | A kind of ZigBee voice controllers | |
| CN109936419A (en) | A kind of unmanned plane broadcast system and method | |
| KR20230110941A (en) | Electronic device and mehtod for determining location using uwb signal in electronic device | |
| WO2021088582A1 (en) | Internet-of-things sensor management method and wireless access point | |
| KR20060107890A (en) | Wireless home network system by speech recognition | |
| KR20220162247A (en) | Voice processing device for processing voice of speaker according to authority level | |
| CN110808042A (en) | Voice interaction networking system and method | |
| CN113270094A (en) | Electrical equipment control method and device | |
| CN106847280B (en) | Audio information processing method, intelligent terminal and voice control terminal |
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: 20200714 |