CN209265517U - A kind of vocal print punch card system based on speech recognition - Google Patents
A kind of vocal print punch card system based on speech recognition Download PDFInfo
- Publication number
- CN209265517U CN209265517U CN201822130739.0U CN201822130739U CN209265517U CN 209265517 U CN209265517 U CN 209265517U CN 201822130739 U CN201822130739 U CN 201822130739U CN 209265517 U CN209265517 U CN 209265517U
- Authority
- CN
- China
- Prior art keywords
- module
- audio
- card
- cpu module
- punch
- 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
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The vocal print punch card system based on speech recognition that the utility model relates to a kind of, specifically include that punch-card device and server, the punch-card device includes: CPU module, audio collection processing module, WIFI module and audio output module, wherein, the audio collection processing module is connected with the CPU module, for by the audio information transmissions after acquisition process to the CPU module；The server is connected by the WIFI module with the CPU module, after obtaining lteral data by audio-frequency information progress speech recognition and carry out natural language processing to the lteral data, the corresponding reply text of the lteral data is sent to the CPU module and handle as audio data；The audio output module is connected with the CPU module, for playing out to the audio data.The utility model reduces hard disk cost for storing audio file and intelligent higher, has better interactive experience.
Description
Technical field
The utility model belongs to attendance and checks card technical field, and in particular to a kind of vocal print based on speech recognition, which is checked card, is
System.
Background technique
The function that major part voice punch-card device or client are realized at present is more single, cannot identify voice content, only
It is capable of fixing the audio file that casting records in advance, intelligence degree is very low, and when quantity of documents in audio file library
When more, it is at high cost especially to waste memory space.Such as the patent document that application number number is 201410736115.7, disclose one kind
The method for playing voice messaging by punched-card machine, the technology are played by computer control and are recorded in advance when someone checks card
Good audio file, cannot recognize that voice content described in the people that checks card to give correct reply, as punched-card machine cannot be known
The people that Chu not check card inquiry is that real-time weather or real-time time either inquire whether some colleague has come to company etc.,
Real-time weather and real-time time and the working information of some corporator etc. can not be replied, it can only mechanically be played
Audio file through recording, simultaneously, if content is more in his sound bank, the memory space of cost
Greatly, intelligent and hardware cost low with humanization level is high.
Utility model content
The utility model in view of the deficiencies of the prior art, proposes a kind of vocal print punch card system based on speech recognition, should
System reply check card people when, server is using character solution, and punch-card device is got after text from turn later
It changes audio into, reduces the hard disk cost for being used to store audio file, while using speech recognition technology and natural language processing
(NLP) algorithm can carry out the specific answer of real-time for particular problem, intelligent higher, have better interactive experience.
In order to solve the above technical problems, the technical solution that the utility model is taken are as follows:
A kind of vocal print punch card system based on speech recognition, comprising: punch-card device and server, the punch-card device packet
It includes: CPU module, audio collection processing module, WIFI module and audio output module, wherein the audio collection processing module
It is connected with the CPU module, for by the audio information transmissions after acquisition process to the CPU module；The server passes through
The WIFI module is connected with the CPU module, and audio-frequency information progress speech recognition is obtained lteral data and to this article
After digital data carries out natural language processing, the corresponding reply text of the lteral data is sent to the CPU module and is handled
For audio data；The audio output module is connected with the CPU module, for playing out to the audio data.
Further, the voice collecting processing module includes: microphone module, MCU module and sound signal processing mould
Block, wherein the microphone module is used to being converted to acoustic signals into voltage signal, the MCU module and the microphone mould
Block is connected, for by the voltage signal acquisition at audio-frequency information, the sound signal processing module respectively with the MCU module
It is connected with the CPU module, for being sent to the CPU module after handling the audio-frequency information.
Further, the server includes: voice recognition unit, check card information memory cell and natural language processing list
Member, wherein the information memory cell of checking card is for storing the information of checking card for the people that checks card.
Further, the audio output module is loudspeaker.
The beneficial effects of the utility model include at least: the utility model reduce the hard disk for storing audio file at
This, while the specific of real-time can be carried out for particular problem using speech recognition technology and natural language processing (NLP) algorithm
It answers, it is intelligent higher, there is better interactive experience.
Detailed description of the invention
Fig. 1 is the utility model vocal print punch card system structural block diagram.
Fig. 2 is the utility model punch card method flow chart.
Specific embodiment
In order to enable those skilled in the art to better understand the technical solutions of the present invention, combined with specific embodiments below
The utility model is described in further detail.The embodiments described below is exemplary, and is only used for explaining that this is practical new
Type, and should not be understood as limiting the present invention.Particular technique or condition are not specified in embodiment, according in the art
Document described in technology or conditions or carried out according to product description.
The utility model is in order to solve the above technical problems, the main utility model design taken are as follows:
Realize that vocal print is checked card using the technology of speech recognition, it is intended to further increase the interactive experience checked card, especially solve
The problem of real time information etc. of having determined cannot make Audio Reply in advance；It distinguishes and the previous particular content that cannot detect voice
The casting for the audio file that can only be recorded in advance is different, and the utility model can be analyzed by intelligence nlp algorithm and be checked card
The voice content of people, and correct corresponding real-time response is provided, replying formation beyond the clouds is written form, and text is sent to
Punch-card device is local, and then local punch-card device broadcasts text conversion at audio: for example this content replied is related to real-time
Correct text answer can be then obtained by network retrieval, then broadcast text conversion at audio in local punch-card device.
If this answer is related to the relevant information of this enterprise, the information that can check card according to company's local data base or server is deposited
Storage unit is replied to form corresponding text, and punch-card device is converted into audio casting etc. later.Because audio be according to text come
It automatically generates, can accurately reply the inquiry of checking card for the people that checks card in real time in this way, substantially increase compared with the prior art artificial
It is intelligent to be used to store hardware cost brought by audio file with reduction.
Embodiment according to the present utility model, Fig. 1 are the utility model vocal print punch card system structural block diagram, referring to Fig.1 institute
Show, the vocal print punch card system described in the utility model based on speech recognition specifically includes that punch-card device and server, described to beat
Card device includes: audio collection processing module, CPU module, WIFI module and audio output module, it is preferred that the audio is defeated
Module is loudspeaker out.
Embodiment according to the present utility model, shown referring to Fig.1, the voice collecting processing module includes: microphone mould
Block, MCU module and voice signal processing module, wherein the microphone module is used to acoustic signals being converted to voltage signal,
The MCU module is connected with the microphone module, is used for the voltage signal acquisition into audio-frequency information, the voice signal
Processing module is connected with the MCU module, for handling the audio-frequency information, such as: at noise signal filtering and power amplifier
Reason etc..
Embodiment according to the present utility model, shown referring to Fig.1, the sound signal processing module and the CPU module
Be connected, for will treated audio information transmissions to the CPU module；The server by the WIFI module with it is described
CPU module is connected, and receives the audio-frequency information that the CPU module sends over and carries out speech recognition to it and obtains text number
According to, at the same to the lteral data carry out natural language processing (NLP) after, the corresponding reply text of the lteral data is sent in real time
Carrying out processing to the CPU module is audio data；The audio output module is connected with the CPU module, for described
Audio data plays out.
Embodiment according to the present utility model, server described in the utility model include: cloud voice print database library unit, voice
Recognition unit, check card information memory cell and natural language processing unit, Fig. 2 are the utility model punch card method flow chart, ginseng
According to shown in Fig. 2, the utility model punch card method specifically includes following steps.
Step 1: the vocal print of corporator is uploaded to the cloud voice print database library unit of server after being registered, as the people that checks card
Wake-up punch-card device (such as: wheat fine jade classmate allows an xx to come Hou Qu manager office, company meeting；For another example: wheat fine jade classmate looks into
Ask the city xx this afternoon weather) when, acquire the cloud voice print database library unit that the speech audio data input is uploaded to server.Root
According to the embodiments of the present invention, the audio-frequency information of the upload is by the processing such as noise signal filtering and power amplifier processing.
Step 2: whether the vocal print for judging current speech is the registered member that checks card, and is, record is currently checked card and people and beaten
The information such as card time are deposited into check card information memory cell or the local data storage module of server, are convenient for company
Record of checking card is transferred at any time.
Step 3: while the audio-frequency information of the acquisition is converted by word character string by voice recognition unit, then pass through institute
It states natural language processing unit and area is carried out by word character string using NLPP algorithm to audio-frequency information progress speech analysis
Section parsing parses with lower curtate and obtains behavior intention.
Step 4: after behavior is intended to acquisition, the corresponding text of analysis generation being carried out to event behavior classification and is replied and behaviour
Make, the text reply of generation is then sent to punch-card device.
It is understood that event is not limited to above-mentioned notice and inquiry, there are also chat greetings, requesting song, prompting etc..
Step 5: the reply text that punch-card device will acquire carries out transcoding generation audio and is broadcasted.
The step 3- step 5 is specific for example: using speech recognition technology, speech audio information is converted into hereafter
Text is carried out section parsing, is parsed with lower curtate and obtain intention by word using NLP algorithm:
Movement: notice inquiry；
Object: xx weather；
Place: the city xx, manager office；
Time: horse back this afternoon.
Notification event behavior is intended to: notice xx arrives manager office at once；The text of transmission reply at the same time " it is good,
Can be in first time notice xx to manager office " punch-card device is arrived, punch-card device carries out text and turns audio casting.
Query event behavior is intended to: the inquiry city xx weather this afternoon obtains the city xx weather this afternoon from network, then
The reply of weather text is transferred to punch-card device, text is converted to audio and broadcasted by punch-card device.
Embodiment according to the present utility model, vocal print punch card system described in the utility model can be used in intelligent sound speaker
Etc. in the electronic products such as correlations.
In conclusion the vocal print punch card system described in the utility model based on speech recognition, reply check card people when,
For server using character solution, punch-card device is automatically converted into audio after getting text, reduces and is used to store audio text
The hard disk cost of part, while can be carried out in fact for particular problem using speech recognition technology and natural language processing (NLP) algorithm
The specific answer of when property, it is intelligent higher, there is better interactive experience.
In the present invention unless specifically defined or limited otherwise, the terms such as term " connected ", " connection " should be done extensively
Reason and good sense solution may be a detachable connection for example, it may be being fixedly connected, or integral；It can be mechanical connection, it can also be with
It is electrical connection；Can be directly connected, can also indirectly connected through an intermediary, can be connection inside two elements or
The interaction relationship of two elements.For the ordinary skill in the art, can understand as the case may be above-mentioned
The concrete meaning of term in the present invention.
Although the embodiments of the present invention have been shown and described above, it is to be understood that above-described embodiment is
Illustratively, it should not be understood as limiting the present invention, those skilled in the art are in the scope of the utility model
It inside can make changes, modifications, alterations, and variations to the above described embodiments, at the same time, for those skilled in the art, foundation
The thought of the application, there will be changes in the specific implementation manner and application range.
Claims (4)
1. a kind of vocal print punch card system based on speech recognition characterized by comprising punch-card device and server, it is described to beat
Card device includes: CPU module, audio collection processing module, WIFI module and audio output module, wherein the audio collection
Processing module is connected with the CPU module, for by the audio information transmissions after acquisition process to the CPU module；
The server is connected by the WIFI module with the CPU module, and audio-frequency information progress speech recognition is obtained
After carrying out natural language processing to lteral data and to the lteral data, the corresponding reply text of the lteral data is sent to institute
Stating CPU module and carrying out processing is audio data；
The audio output module is connected with the CPU module, for playing out to the audio data.
2. vocal print punch card system according to claim 1, which is characterized in that the voice collecting processing module includes: wheat
Gram wind module, MCU module and voice signal processing module, wherein the microphone module is used to acoustic signals being converted to electricity
Signal is pressed, the MCU module is connected with the microphone module, is used for the voltage signal acquisition into audio-frequency information, described
Sound signal processing module is connected with the MCU module and the CPU module respectively, for handling the audio-frequency information
After be sent to the CPU module.
3. vocal print punch card system according to claim 1, which is characterized in that the server include: voice recognition unit,
It checks card information memory cell and natural language processing unit, wherein the information memory cell of checking card checks card people's for storing
It checks card information.
4. vocal print punch card system according to claim 1, which is characterized in that the audio output module is loudspeaker.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201822130739.0U CN209265517U (en) | 2018-12-18 | 2018-12-18 | A kind of vocal print punch card system based on speech recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201822130739.0U CN209265517U (en) | 2018-12-18 | 2018-12-18 | A kind of vocal print punch card system based on speech recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN209265517U true CN209265517U (en) | 2019-08-16 |
Family
ID=67560142
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201822130739.0U Active CN209265517U (en) | 2018-12-18 | 2018-12-18 | A kind of vocal print punch card system based on speech recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN209265517U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112542156A (en) * | 2020-12-08 | 2021-03-23 | 山东航空股份有限公司 | Civil aviation maintenance worker card system based on voiceprint recognition and voice instruction control |
| CN116071842A (en) * | 2022-11-21 | 2023-05-05 | 深聪半导体(江苏)有限公司 | Voiceprint-based card punching method and system |
-
2018
- 2018-12-18 CN CN201822130739.0U patent/CN209265517U/en active Active
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112542156A (en) * | 2020-12-08 | 2021-03-23 | 山东航空股份有限公司 | Civil aviation maintenance worker card system based on voiceprint recognition and voice instruction control |
| CN116071842A (en) * | 2022-11-21 | 2023-05-05 | 深聪半导体(江苏)有限公司 | Voiceprint-based card punching method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103929537B (en) | Real-time reminding method based on messages of different levels | |
| JP2020525903A (en) | Managing Privilege by Speaking for Voice Assistant System | |
| WO2019227580A1 (en) | Voice recognition method, apparatus, computer device, and storage medium | |
| CN111488433A (en) | Artificial intelligence interactive system suitable for bank and capable of improving field experience | |
| CN101262524A (en) | Rubbish voice filtration method and system | |
| CN108766428A (en) | A kind of voice broadcast control method and system | |
| CN103886857B (en) | A kind of noise control method and equipment | |
| CN102497391B (en) | Server and reminding method | |
| CN209265517U (en) | A kind of vocal print punch card system based on speech recognition | |
| CN109256136A (en) | A kind of audio recognition method and device | |
| CN103093752A (en) | Sentiment analytical method based on mobile phone voices and sentiment analytical system based on mobile phone voices | |
| CN105244042B (en) | A kind of speech emotional interactive device and method based on finite-state automata | |
| CN102404278A (en) | Song request system based on voiceprint recognition and application method thereof | |
| CN107645523A (en) | A kind of method and system of mood interaction | |
| CN103546503A (en) | Voice-based cloud social system, voice-based cloud social method and cloud analysis server | |
| CN108899036A (en) | A kind of processing method and processing device of voice data | |
| CN111145763A (en) | GRU-based voice recognition method and system in audio | |
| CN104702759A (en) | Address list setting method and address list setting device | |
| CN101867742A (en) | Television system based on sound control | |
| CN109274556A (en) | A kind of collection and analysis system of web log | |
| CN112102846A (en) | Audio processing method and device, electronic equipment and storage medium | |
| CN110415678A (en) | Customized voice broadcast client, server, system and method | |
| CN109686360A (en) | A kind of voice is made a reservation robot | |
| CN107358958B (en) | Intercommunication method, apparatus and system | |
| CN102347845A (en) | Broadcasting system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |