CN208284231U - Voice enhancer based on voice awakening technology - Google Patents
Voice enhancer based on voice awakening technology Download PDFInfo
- Publication number
- CN208284231U CN208284231U CN201820590073.4U CN201820590073U CN208284231U CN 208284231 U CN208284231 U CN 208284231U CN 201820590073 U CN201820590073 U CN 201820590073U CN 208284231 U CN208284231 U CN 208284231U
- Authority
- CN
- China
- Prior art keywords
- voice
- module
- signal
- input terminal
- output end
- 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.)
- Active
Links
Landscapes
- Circuit For Audible Band Transducer (AREA)
Abstract
The utility model discloses a kind of voice enhancer based on voice awakening technology, the microphone array including being made of more than two microphones, the voice for picking up the microphone array carry out the amplification module of signal enhanced processing, are used to for the amplified voice signal being sampled the Dynamic Signal sampling module of processing；It further include that voice signal for sampling Dynamic Signal carries out that the linear differences processing module of setting sound frame, be used for will be by linear differences treated adaptive-filtering module that voice signal is filtered, the speech enhan-cement module for being used to enhance filtered voice signal, the TDOA computing module and network communication module for being used to determine sound source position.The utility model can turn off or on corresponding household appliance according to sound source position, avoid traditional voice control position not accurately defect, the intelligent level of intelligent home voice interactive system is made to get a promotion.
Description
Technical field
The utility model relates to smart home fields, more particularly to a kind of speech enhan-cement based on voice awakening technology
Device.
Background technique
With the development of science and technology, smart home is gradually generalized application, and in the implementation process of smart home, interactive voice
To realize the intelligentized important technical of smart home, the voice interactive system of smart home at present, wake up language and
In the pick process of order language, external environment noise and the interference that other people speak inevitably will receive, if interference is made an uproar
Sound is too strong, can seriously affect the sensitivity of voice interactive system, causes to wake up failure and interactive voice is unsmooth；Moreover, because
Indoor household appliances installation site is more dispersed, existing interactive voice cannot very accurately according to the position of sounder open or
The problems such as closing household appliance corresponding with position of human body, there is mistakes to control for voice control, mixing.
Summary of the invention
The purpose of the utility model is to overcome shortcoming in the prior art, provide a kind of based on voice awakening technology
Voice enhancer can improve the interaction of language smart home middle pitch according to the corresponding household appliance of position of human body voice control
Accuracy and sensitivity.
In order to solve the above technical problems, the utility model is solved by following technical proposals:
A kind of voice enhancer based on voice awakening technology, including
Corpus sampling mold group, voice pre-treatment mould group and voice post-process mould group；
The corpus sampling mold group includes the microphone array being made of more than two microphones, is used for the microphone
The voice that array picks up carries out the amplification module of signal enhanced processing and for the amplified voice signal to be sampled
The Dynamic Signal sampling module of processing；
The voice pre-treatment mould group includes the linear differences for the voice signal of Dynamic sampling to be carried out to setting sound frame
Processing module, treated for will pass through linear differences, and voice signal carries out the filter module of adaptive-filtering processing and is used for
Enhance the speech enhan-cement module of filtered voice signal；
Voice post-processing mould group includes TDOA computing module for determining sound source position and for by the source of sound
Location information and voice signal are sent to the network communication module of intelligent terminal；
The output end of the microphone array connects the input terminal of the amplification module, and the output end of the amplification module connects
The input terminal of Dynamic Signal sampling module is connect, the output end of the Dynamic sampling module connects the linear differences processing module
Input terminal, the output end of the linear differences processing module connect the input terminal of the filter module, the filter module it is defeated
Outlet connects the input terminal of the speech enhan-cement module, and the output end of the speech enhan-cement module connects the TDOA simultaneously and calculates
The output end of the input terminal of module and the input terminal of network communication module, the TDOA computing module is also connected with the network communication
The input terminal of module, the output end of the network communication module connect the input terminal of the intelligent terminal.
Further, described two above microphones be distributed in the different location in room and with the voice enhancer
It is electrically connected.
Further, the linear differences processing module is also used to be cut into the voice signal setting volume limit value of sound frame.
Further, described two above microphones are detachably connected with the voice enhancer by interface.
Further, the network communication module carries out wireless signal transmission by ZIGBEE technology.
The utility model have the following advantages that compared with prior art and the utility model has the advantages that
(1) the utility model by setting indoors different location more than two microphones composition microphone array into
The pickup of row voice, by amplifying, being superimposed to the voice that microphone array picks up, enhancing processing, effectively to original language
Sound has carried out noise suppressed and reverberation is eliminated, and improves indoor different location and the speech recognition capabilities compared with amount of bass.
(2) the Dynamic Signal sampling module of the utility model is by phonetic sampling algorithm accurately in the voice environment mixed
Middle extraction target voice information, improves the speech recognition capabilities in interfering noise environment, and the linear differences processing module is logical
The volume limit value for crossing setting filters out the lesser part of target voice volume, avoids a sounding position to other compared with distant positions man
The voice of electric equipment interferes, and keeps voice control position more accurate.
(3) the TDOA computing module obtains the incident deflection of source of sound according to deflection algorithm, passes through the direction
Angle determines sound source position, then module is sent to intelligent end by wireless communication by the sound source position and enhancing audio digital signals
End, intelligent terminal can open or close near sound source position according to the sound source position information and enhancing audio digital signals
Household appliance, such design make voice control more agree with the actual wishes of people, improve the intelligent experience of interactive voice.
(4) multiple microphones of the utility model are electrically connected by interface with the voice enhancer, by setting indoors
The microphone and respective array structure of different number are set, can reasonably be laid out microphone according to the size in room, space structure
Array structure, optimizes resource distribution, flexibly improves the pickup effect and speech enhan-cement effect of microphone array.
Detailed description of the invention
Fig. 1 is the principle flow chart of voice enhancer of the utility model based on voice awakening technology.
Specific embodiment
The present invention will be further described in detail with reference to the embodiments and the accompanying drawings, but the implementation of the utility model
Mode is without being limited thereto.
As shown in Figure 1, a kind of voice enhancer based on voice awakening technology, comprising: before corpus sampling mold group 1, voice
It handles mould group 2 and voice post-processes mould group 3.
The corpus sampling mold group includes the microphone array 4 being made of more than two microphones, is used for the Mike
The voice that wind array picks up carries out the amplification module 5 of signal enhanced processing and for taking the amplified voice signal
The Dynamic Signal sampling mold 6 of sample processing；The voice pre-treatment mould group 2 include voice signal for sampling Dynamic Signal into
The linear differences processing module 7 of row setting sound frame, treated for will pass through linear differences, and voice signal is adaptively filtered
The filter module 8 that wave is handled and the speech enhan-cement module 9 for enhancing filtered voice signal；The voice post-processes mould group
Including for determining the TDOA computing module 10 of sound source position and for the sound source position information and voice signal to be sent to
The network communication module 11 of intelligent terminal.
The output end of the microphone array 4 connects the input terminal of the amplification module 5, the output of the amplification module 5
The input terminal of end connection Dynamic Signal sampling module 6, the output end of the Dynamic sampling module 6 connect the linear differences processing
The input terminal of module 7, the output end of the linear differences processing module 7 connect the input terminal of the filter module 8, the filtering
The output end of module 8 connects the input terminal of the speech enhan-cement module 9, and the output end of the speech enhan-cement module 9 connects simultaneously
The input terminal of the TDOA computing module 10 and the input terminal of network communication module 11, the output end of the TDOA computing module 10
It is also connected with the input terminal of the network communication module 11, the output end of the network communication module 11 connects the intelligent terminal 12
Input terminal.
The working principle of the present embodiment is as follows: it is to utilize microphone that the microphone array 4 of the present embodiment, which carries out source of sound positioning,
Array 4 receives voice signal, the locality of source of sound is judged, mainly using same source of sound into microphone array 4 every
The difference of the distance of microphone, therefore the voice signal of same source of sound is transmitted to each microphone and has time difference TDOA, benefit
With the TDOA acquired, substituting into deflection algorithmic formula can be obtained the direction of source of sound incidence.Language is received by microphone array 4 first
Sound signal, the received voice signal of microphone is after the amplification of amplification module 5, then via Dynamic Signal sampling module 6, according to setting
Sampling frequency sampled voice signal and analyzed, the linear differences processing module 7 is by the received voice of every microphone
Signal volume standardizes and is cut into multiple sound frames, then the voice signal for being cut into sound frame is set volume limit value, removal volume religion
Small part calculates the volume of all sound frames in one section of voice signal, then is believed the voice of all sound frame volumes by filter module 8
Number adaptive-filtering processing is carried out, obtains adjusting audio digital signals, the speech enhan-cement module 9 is by the tune of the multiple sound frame
Section audio digital signals are overlapped processing, generate enhancing audio digital signals, meanwhile, the microphone array is estimated according to algorithm
The TDOA acquired is substituted into deflection algorithmic formula, the deflection of source of sound incidence can be obtained by the TDOA of every microphone in column,
The calculated deflection data-signal of TDOA module and enhancing audio digital signals are sent to intelligence by the network communication module 3
Terminal 12, the intelligent terminal 12 open or close diaphone according to the deflection data-signal and enhancing audio digital signals
The household appliance of source position carries out the position that source of sound positions determination sounder that can be more accurate using the positioning method of TDOA
Information, even if sounder position can also be accurately positioned within the scope of smaller space length.The utility model passes through as a result,
It determines sound source position, and corresponding household appliance is turned off or on according to sound source position, avoid traditional voice control position not
Accurately defect, voice control preferably meet people's actual demand, make the intelligent water of intelligent home voice interactive system
It is flat to get a promotion.
According to technical solutions of the utility model, described two above microphones be distributed in the different location in room and with
The voice enhancer is electrically connected, since in sound source, from the farther away situation of microphone, the signal that microphone receives is often
It is interfered by reverberation caused by range attenuation, noise jamming and echo, leads to the decline of voice quality, the present embodiment passes through in room
Interior suitable location arrangements microphone array array structure, efficiently solves the above problem, is directed toward sound source by multichannel microphone and carries out
Wave beam forming improves speech recognition capabilities to obtain the source of sound input of high quality.
Described two above microphones are detachably connected with the voice enhancer by interface, and the voice enhancer can
With it is embedded or it is concealed be installed on indoor wall, different number can be arranged indoors by being detachably connected by interface
Microphone simultaneously sets respective array structure to the microphone of different number, can be according to the size in room, and space structure is reasonable
It is laid out microphone array array structure, resource distribution is optimized, is flexibly improved in not chummery by different array structures
The pickup effect and speech enhan-cement effect of microphone array.
The network communication module carries out wireless signal transmission by ZIGBEE technology, using ZIGBEE network communication mode
So that intelligent terminal is connected simultaneously and control the greater number of voice enhancer, and keeps good signal transmission matter
Amount, reduces power consumption, saves material cost.
Above-described embodiment is the preferable embodiment of the utility model, but the embodiments of the present invention is not by above-mentioned
The limitation of embodiment, it is made under other any spiritual essence and principles without departing from the utility model to change, modify, replacing
In generation, simplifies combination, should be equivalent substitute mode, is included within the protection scope of the utility model.
Claims (5)
1. a kind of voice enhancer based on voice awakening technology, it is characterised in that: including
Corpus sampling mold group, voice pre-treatment mould group and voice post-process mould group；
The corpus sampling mold group includes the microphone array being made of more than two microphones, is used for the microphone array
The voice of pickup carries out the amplification module of signal enhanced processing and for the amplified voice signal to be sampled processing
Dynamic Signal sampling module；
The voice pre-treatment mould group includes the linear differences that the voice signal for sampling Dynamic Signal carries out setting sound frame
Processing module, treated for will pass through linear differences, and voice signal carries out the filter module of adaptive-filtering processing and is used for
Enhance the speech enhan-cement module of filtered voice signal；
Voice post-processing mould group includes TDOA computing module for determining sound source position and for by the sound source position
Information and voice signal are sent to the network communication module of intelligent terminal；
The output end of the microphone array connects the input terminal of the amplification module, and the output end connection of the amplification module is dynamic
The input terminal of state sample of signal module, the output end of the Dynamic sampling module connect the input of the linear differences processing module
End, the output end of the linear differences processing module connect the input terminal of the filter module, the output end of the filter module
The input terminal of the speech enhan-cement module is connected, the output end of the speech enhan-cement module connects the TDOA computing module simultaneously
Input terminal and network communication module input terminal, the output end of the TDOA computing module is also connected with the network communication module
Input terminal, the output end of the network communication module connects the input terminal of the intelligent terminal.
2. the voice enhancer according to claim 1 based on voice awakening technology, it is characterised in that: more than described two
Microphone is distributed in the different location in room and is electrically connected with the voice enhancer.
3. the voice enhancer according to claim 1 based on voice awakening technology, it is characterised in that: the linear differences
Processing module is also used to be cut into the voice signal setting volume limit value of sound frame.
4. the voice enhancer according to claim 1 based on voice awakening technology, it is characterised in that: more than described two
Microphone is detachably connected with the voice enhancer by interface.
5. the voice enhancer according to claim 1 based on voice awakening technology, it is characterised in that: the network communication
Module carries out wireless signal transmission by ZIGBEE technology.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201820590073.4U CN208284231U (en) | 2018-04-24 | 2018-04-24 | Voice enhancer based on voice awakening technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201820590073.4U CN208284231U (en) | 2018-04-24 | 2018-04-24 | Voice enhancer based on voice awakening technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN208284231U true CN208284231U (en) | 2018-12-25 |
Family
ID=64751015
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201820590073.4U Active CN208284231U (en) | 2018-04-24 | 2018-04-24 | Voice enhancer based on voice awakening technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN208284231U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108389587A (en) * | 2018-04-24 | 2018-08-10 | 苏州宏云智能科技有限公司 | Voice enhancer based on voice awakening technology |
-
2018
- 2018-04-24 CN CN201820590073.4U patent/CN208284231U/en active Active
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108389587A (en) * | 2018-04-24 | 2018-08-10 | 苏州宏云智能科技有限公司 | Voice enhancer based on voice awakening technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102164328B (en) | Audio input system used in home environment based on microphone array | |
| CN106898348B (en) | Dereverberation control method and device for sound production equipment | |
| CN110648678B (en) | Scene identification method and system for conference with multiple microphones | |
| CN108389587A (en) | Voice enhancer based on voice awakening technology | |
| CN105825864B (en) | Both-end based on zero-crossing rate index is spoken detection and echo cancel method | |
| WO2017218128A1 (en) | Far field automatic speech recognition pre-processing | |
| CN100535992C (en) | Small scale microphone array speech enhancement system and method | |
| CN104810021B (en) | The pre-treating method and device recognized applied to far field | |
| CN110010147A (en) | A kind of method and system of Microphone Array Speech enhancing | |
| CN101819758B (en) | System of controlling screen display by voice and implementation method | |
| JP2021110938A (en) | Multiple sound source tracking and speech section detection for planar microphone array | |
| CN106782584A (en) | Audio signal processing apparatus, method and electronic equipment | |
| KR20130084298A (en) | Systems, methods, apparatus, and computer-readable media for far-field multi-source tracking and separation | |
| CN108320749A (en) | Far field voice control device and far field speech control system | |
| CN105679329A (en) | Microphone array voice enhancing device adaptable to strong background noise | |
| CN107124647A (en) | A kind of panoramic video automatically generates the method and device of subtitle file when recording | |
| CN205754809U (en) | A kind of robot self-adapting volume control system | |
| AU2020323929A1 (en) | Acoustic echo cancellation control for distributed audio devices | |
| CN106992010A (en) | Without the microphone array speech enhancement device under the conditions of direct sound wave | |
| CN110517704A (en) | A kind of speech processing system based on microphone array beamforming algorithm | |
| CN208284231U (en) | Voice enhancer based on voice awakening technology | |
| CN106340306A (en) | Method and device for improving speech recognition degree | |
| CN111883161A (en) | Method and device for audio acquisition and position identification | |
| CN103117083A (en) | Audio information acquisition device and method | |
| CN208367199U (en) | Separate type microphone array |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |