CN204066721U - High precision speech recognition device - Google Patents
High precision speech recognition device Download PDFInfo
- Publication number
- CN204066721U CN204066721U CN201420493931.5U CN201420493931U CN204066721U CN 204066721 U CN204066721 U CN 204066721U CN 201420493931 U CN201420493931 U CN 201420493931U CN 204066721 U CN204066721 U CN 204066721U
- Authority
- CN
- China
- Prior art keywords
- feedforward
- rear portion
- wave filter
- data processing
- mcu
- 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.)
- Expired - Fee Related
Links
Landscapes
- Circuit For Audible Band Transducer (AREA)
Abstract
High precision speech recognition device, comprise three sound pick up equipments, each sound pick up equipment rear portion is all connected with independently wave filter, and described wave filter rear portion has been arranged in parallel two groups of data processing equipments, and connect MCU at two groups of data processing equipment rear portions, and by MCU at rear end connection device electric power management circuit, one group in described two groups of data processing equipments comprises a Feedforward Decoupling device, and be in series with the audio process of an integrated form at this feedforward decoupler rear portion, the direct-connected wave filter of this feedforward decoupler front-end interface; Another group data processing equipment in be then only provided with an independently Feedforward Decoupling device, and wave filter rear portion interconnected between two and by this feedforward decoupler be connected to MCU.The utility model control accuracy is high, easy to implement and input is not high, is applicable to very much promoting the use of.
Description
Technical field
The utility model belongs to electronic sensor and control technology field, is specifically related to the high precision speech recognition device in a kind of spoken digit recognition technology.
Background technology
Phonetic entry technology is considered to the Main way of man-machine interaction development, and increasing equipment adopts speech recognition system to carry out text event detection, Voice command and interactive voice.And speech recognition device is as the core of speech recognition system, mainly identification collection is carried out to voice, and the voice signal after gathering is sent into voice transfer module and carried out Signal transmissions after coded treatment.Due to, existing voice identification control device, general is all man-to-manly independently control product, in noisy environment, unsharp pronunciation and many people pronunciation situation, and phonetic entry weak effect, discrimination is low, even None-identified, and easily produce interference between different phonic symbols, do not form real system interlink function, thus also more single in application aspect, the actual needs of modern thing connection environment cannot be met.
And now, input equipment is replaced to carry out the embodiment of function control as the hommization of human-computer interaction function by speech recognition, speech identifying function is made to have become mainstream configuration in many field of intelligent control, therefore, the height of phonetic recognization rate directly has influence on the experience of user, and then directly has influence on the market competitiveness of Related product.
Utility model content
The technical matters that the utility model solves is to provide a kind of high precision speech recognition device, to solve the defect in above-mentioned technical background.
The technical matters that the utility model solves realizes by the following technical solutions:
High precision speech recognition device, comprise three sound pick up equipments, each sound pick up equipment rear portion is all connected with independently wave filter, and described wave filter rear portion has been arranged in parallel two groups of data processing equipments, and connect MCU at two groups of data processing equipment rear portions, and by MCU at rear end connection device electric power management circuit, one group in described two groups of data processing equipments comprises a Feedforward Decoupling device, and be in series with the audio process of an integrated form at this feedforward decoupler rear portion, the direct-connected wave filter of this feedforward decoupler front-end interface; Another group data processing equipment in be then only provided with an independently Feedforward Decoupling device, and wave filter rear portion interconnected between two and by this feedforward decoupler be connected to MCU.
In the utility model, described sound pick up equipment preferably adopts contact sound channel vibration transducer.
In the utility model, described MCU is connected by RS-485 interface with between electric power management circuit.
In the utility model, in described MCU, built-in anti-deadlock house dog treatment circuit is to improve security of system grade.
In the utility model, described electric power management circuit rear portion is also provided with the audio parameter contrast detector of a band display device, in order to show pickup file in real time, parameter in control object is shown, and conveniently the Output rusults of speech recognition device is manually revised, make whole system carry out human-computer interaction.
Beneficial effect: the utility model by increasing the number of sound pick up equipment in pickup, change traditional unidirectional pickup processing mode is the complementary pickup mode of multi-source simultaneously, coordinate audio frequency processing circuit, the extraneous impact interfered voice messaging collection can be greatly reduced, greatly improve its phonetic recognization rate.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of the utility model preferred embodiment.
Wherein: 1, the first sound pick up equipment; 2, the second sound pick up equipment; 3, the 3rd sound pick up equipment; 4, the first wave filter; 5, the second wave filter; 6, the 3rd wave filter; 7, the first Feedforward Decoupling device; 8, audio frequency processing circuit; 9, the second Feedforward Decoupling device; 10, MCU; 11, electric power management circuit; 12, audio parameter contrast detector.
Embodiment
The technological means realized to make the utility model, creation characteristic, reaching object and effect is easy to understand, below in conjunction with concrete diagram, setting forth the utility model further.
In preferred embodiment see a kind of high precision speech recognition device of Fig. 1, in the present embodiment, comprise the first sound pick up equipment 1, second sound pick up equipment 2, 3rd sound pick up equipment 3 three sound pick up equipment, these three sound pick up equipments are contact sound channel vibration transducer, be directly installed on pronunciation unit, in the present embodiment, puberty unit is pickup microphone, three pickup units are evenly arranged in the form of a ring on microphone, its rear section does not connect the first wave filter 4, second wave filter 5, 3rd wave filter 6, and the first wave filter 4, second wave filter 5, 3rd wave filter 6 connects the first Feedforward Decoupling device 7 at rear portion, and connect audio frequency processing circuit 8 at the first Feedforward Decoupling device 7 rear portion, be connected with the MCU10 as main control unit by audio frequency processing circuit 8, in addition, at the first wave filter 4, second wave filter 5, interconnected between two and connect the second Feedforward Decoupling device 9 at rear portion between 3rd wave filter 6, the second direct-connected MCU10 in Feedforward Decoupling device 9 rear portion, its resolution data as a comparison data of data to audio frequency processing circuit 8 filters to know signal after exporting parsing.
And electric power management circuit 11 is connected with at MCU10 rear portion, electric power management circuit 11 is provided with the power input interface of distinct device, equipment controls to realize the switch control rule to distinct device to electric power management circuit 11 by MCU10, the audio parameter contrast detector 12 of a band display device is then connected with at electric power management circuit 11 rear portion, in order to show pickup file in real time, parameter in control object is shown, and conveniently the Output rusults of speech recognition device is manually revised, make whole system carry out human-computer interaction.
More than show and describe ultimate principle of the present utility model, principal character and advantage of the present utility model.The technician of the industry should understand; the utility model is not restricted to the described embodiments; what describe in above-described embodiment and instructions just illustrates principle of the present utility model; under the prerequisite not departing from the utility model spirit and scope; the utility model also has various changes and modifications, and these changes and improvements all fall within the scope of claimed the utility model.The claimed scope of the utility model is defined by appending claims and equivalent thereof.
Claims (4)
1. high precision speech recognition device, it is characterized in that, comprise three sound pick up equipments, each sound pick up equipment rear portion is all connected with independently wave filter, and described wave filter rear portion has been arranged in parallel two groups of data processing equipments, and connect MCU at two groups of data processing equipment rear portions, and by MCU at rear end connection device electric power management circuit, one group in described two groups of data processing equipments comprises a Feedforward Decoupling device, and be in series with the audio process of an integrated form at this feedforward decoupler rear portion, the direct-connected wave filter of this feedforward decoupler front-end interface; Another group data processing equipment in be then only provided with an independently Feedforward Decoupling device, and wave filter rear portion interconnected between two and by this feedforward decoupler be connected to MCU.
2. a kind of high precision speech recognition device according to claim 1, is characterized in that, described sound pick up equipment preferably adopts contact sound channel vibration transducer.
3. a kind of high precision speech recognition device according to claim 1, it is characterized in that, described MCU is connected by RS-485 interface with between electric power management circuit.
4. a kind of high precision speech recognition device according to claim 1, is characterized in that, described electric power management circuit rear portion is also provided with the audio parameter contrast detector of a band display device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201420493931.5U CN204066721U (en) | 2014-08-29 | 2014-08-29 | High precision speech recognition device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201420493931.5U CN204066721U (en) | 2014-08-29 | 2014-08-29 | High precision speech recognition device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN204066721U true CN204066721U (en) | 2014-12-31 |
Family
ID=52208181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201420493931.5U Expired - Fee Related CN204066721U (en) | 2014-08-29 | 2014-08-29 | High precision speech recognition device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN204066721U (en) |
-
2014
- 2014-08-29 CN CN201420493931.5U patent/CN204066721U/en not_active Expired - Fee Related
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106340309A (en) | Dog bark emotion recognition method and device based on deep learning | |
| CN103971681A (en) | Voice recognition method and system | |
| CN102981615B (en) | Gesture identifying device and recognition methods | |
| CN103198829A (en) | Method, device and equipment of reducing interior noise and improving voice recognition rate | |
| CN108111538A (en) | Smart projector speech control system and its method based on sound groove recognition technology in e | |
| CN104123930A (en) | Guttural identification method and device | |
| CN202796043U (en) | Voice recognition system | |
| CN104282303A (en) | Method for conducting voice recognition by voiceprint recognition and electronic device thereof | |
| CN104112453A (en) | Audio preprocessing system | |
| CN103117083A (en) | Audio information acquisition device and method | |
| CN202711445U (en) | Digital multimedia remote control | |
| CN204066721U (en) | High precision speech recognition device | |
| CN208000464U (en) | A kind of intelligent sound robot adjusting volume based on the age | |
| CN202471229U (en) | Electronic scale for blind people | |
| CN203911924U (en) | Bluetooth device with voice wake-up | |
| CN104766610A (en) | Voice recognition system and method based on vibration | |
| CN206313861U (en) | A kind of audio detection system based on the control of NI boards serial ports | |
| CN104078042A (en) | Electronic device and information processing method | |
| CN106653040A (en) | Voice audio signal sampling processing method | |
| CN205415318U (en) | Bearing grinding machine real -time monitoring apparatus based on acoustic emission principle | |
| CN205911008U (en) | Many angle location system of intelligent robot | |
| CN206758125U (en) | A kind of application on voiceprint recognition equipment under noise circumstance | |
| CN208925781U (en) | A kind of intelligent sound electrically driven curtain | |
| CN207833286U (en) | A kind of IEPE collecting sensor signals card | |
| CN203414390U (en) | DSP Builder-based blood cell pulse signal detection and statistics device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20141231 Termination date: 20160829 |