CN1452159A - Speech controlling device and method - Google Patents
Speech controlling device and method Download PDFInfo
- Publication number
- CN1452159A CN1452159A CN02114534A CN02114534A CN1452159A CN 1452159 A CN1452159 A CN 1452159A CN 02114534 A CN02114534 A CN 02114534A CN 02114534 A CN02114534 A CN 02114534A CN 1452159 A CN1452159 A CN 1452159A
- Authority
- CN
- China
- Prior art keywords
- voice
- circuit
- phonetic
- output
- control
- 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
Images
Landscapes
- Input From Keyboards Or The Like (AREA)
Abstract
A speech controller and method are disclosed. Its controller is characterized by that an electronic switch is electrically connected to I/O port and its filter has no aliasing distortion. Said method includes preweighting and end-point detection for removing personal difference and environment noise, extracting acoustic parameters, forming speech pattern from said acoustic parameters, and comparing it with prestored reference patterns for finding out a match. Its advantages are simple structure, low cost and high speed recognition efficiency.
Description
Affiliated field: the present invention relates to a kind of voice control multi-channel electronic switch, it can be adapted to the voice control of various household electrical appliance.
Background technology: existing household electrical appliance polygamy has control device of wireless, is used to control the various functions of household electrical appliance, is used to control bright, the colourity of channel, sound of televisor as the telepilot of televisor outfit.Telepilot control wind direction in the air-conditioning, temperature are adjusted or the like.But, but can not be convenient to use this telepilot to the patient of trick activity inconvenience.Though the employing speech recognition system is arranged now, because price height and complexity also do not spread to household electrical appliance.And simple voice control exists the low problem of discrimination.
Chinese patent discloses a kind of＜intercom acoustic control telephone transmitter 〉, its patent No. is 00254108.4, it is to realize that by intercom acoustic control telephone transmitter when signal import, mobile phone was launched signal.
Summary of the invention: the objective of the invention is to design a kind of simple, price is low, phonetic controller that phonetic recognization rate is high and method.
Summary of the invention: the objective of the invention is to design a kind of simple, price is low, phonetic controller that phonetic recognization rate is high and method.
The object of the present invention is achieved like this, designs a kind of phonetic controller,
Which comprises at least: microphone: be used to receive phonetic entry, and change voice input signal into electric signal;
Filtering circuit: the audio-frequency information that is used for a road above-mentioned electric signal is produced simultaneously the output different frequency bands;
The treatment circuit of one band A/D converter is electrically connected with the output of filtering circuit, is used for different frequency bands output changing into the digital quantity storage with filtering circuit;
The digital quantity that is used for and has stored compares, to determine the validity of phonetic entry, and when the voice messaging that reads is effective, the output terminal action of a correspondence of control;
Memory is connected electrically on the data bus of control circuit, is used to store effective speech input information;
Keyboard: be connected electrically in the I/O mouth of control circuit, comprise that at least a button is used to control phonetic entry and whether deposits in the memory; At least also comprise that a button is used for determining that an output port of control circuit is corresponding with a speech input information, when the phonetic entry of correspondence, control circuit moves corresponding output end.
Voice control method comprises two processes of the initialization of voice messaging and voice control output:
1, the initialization of voice messaging, the initialization of voice messaging is that the voice through confirming are transformed into electric signal through microphone, described electric signal is after amplifying circuit amplifies, obtain the electrical information of same audio-frequency information different frequency bands again through the filtering circuit filtering of a plurality of different frequency bands, the electrical information of these different frequency bands changes into corresponding number of bands amount information storage at a certain address field of memory by the processing circuit processes that has multi-channel a/d converter, and described affirmation process is to finish after being confirmed by the acknowledgement key on the keyboard; Memory can deposit the input of a plurality of different phonetic in, and deposits it in different memory address section, and these different memory address sections have determined different control circuit output;
2, voice control output, after the initialization acknowledgement key is finished, enter voice control output, after initialization finished in the voice of confirming, so long as import once more through the voice of confirming, treatment circuit compares with digital quantity in having memory digitized phonetic entry by different addresses, when unanimity, corresponding output terminal is moved.
Advantage of the present invention is: because the present invention adopts the filtering circuit of multichannel different frequency range to extract the phonetic feature of different phonetic in system, deposit in the memory by the treatment circuit that has multi-channel a/d converter and with it, it can effectively remove the noise effect that individual pronunciation difference and environment cause, therefore the phonetic recognization rate height improves the reliability of control system greatly.The treatment circuit that employing has multi-channel a/d converter is that microprocessor carries out speech pattern recognition, and it is low to have a cost, and characteristics simple in structure can be adapted to various electrical equipment.This system can embed in the household electrical appliances such as air-conditioning, light fixture, realizes the voice control of household electrical appliances, thereby substitutes traditional telepilot.
The invention will be further described for knot embodiment accompanying drawing below:
Description of drawings: Fig. 1 is a schematic block circuit diagram of the present invention;
Among the figure, 1, microphone; 2, amplifying circuit; 3, multichannel filtering circuit; 4, the control circuit that has multi-channel a/d converter; 5, keyboard; 6, memory; 7, electronic switch.
Embodiment: as shown in Figure 1, Fig. 1 adopts the embodiment of AT89S8535 integrated circuit as control circuit.It comprises, microphone 1, amplifying circuit 2, multichannel filtering circuit 3, the control circuit 4, memory 6, keyboard 5, the electronic switch 7 that have multi-channel a/d converter are formed, microphone 1 can be selected the good microphone of range, the input of microphone 1 is after the amplifying circuit 2 of low distortion amplifies, be input to the input end of multichannel filtering circuit 3, multichannel filtering circuit 3 can be chosen 3 to 8 audio sections; Actual conditions can liken to select according to performance and price.The control circuit 4 that has multi-channel a/d converter can adopt the AT89S8535 integrated circuit, also can select the integrated circuit of other band A/D converter; Be connected with keyboard 5 on the I/O mouth of the integrated circuit of being with A/D converter, be electrically connected with memory 6 on data bus, be electrically connected with electronic switch 7 on the I/O mouth, electronic switch 7 can replace the button of existing household electrical appliance.During work, by microphone 1 input one voice, when these voice had deposited memory 6 in, the control circuit 4 that has a multi-channel a/d converter was through confirming when correct; Can control electronic switch gating mistake, electronic switch control household electrical appliance are finished a certain function, establish the intensification or the cooling of transferring in the device as sky.As channel selection in the televisor etc.Certainly, this is to realize by control circuit 4 inside programming that have multi-channel a/d converter.
Work of the present invention mainly comprises voice messaging initialization and two processes of voice control output:
The initialization of voice messaging in this program, is pressed voice key is set, and makes control circuit be in the reception voice status; This moment is by microphone input and the corresponding to voice of order; After the control circuit identification, the voice of importing are deposited into memory 4000 to 47FF; 4000 is that the 1st filtering circuit exported pairing address, and 47FF is that the 8th filtering circuit exported pairing address; Invalid as phonetic entry, can carry out mandatum cassatorium; Reenter start program again; Effectively can be as phonetic entry by continuing key, program enters the typing of second command, and the typing address of second command can be stored in memory 4800 to 4FFF; 4800 export pairing address for the 1st filtering circuit of second voice, and 4FFF is that the 8th filtering circuit exported pairing address; Invalid as phonetic entry, can carry out mandatum cassatorium; Re-execute above program again; How many bar orders phonetic entry has to decide according to system design, and its working routine is identical; Press end key the finish command at last, above-mentioned declarative procedure just for understand convenient for example, actual design is to carry out structural design because of concrete circuit.
Voice control output, after the voice initialization procedure was finished, phonetic controller had just had the ability of recognizing voice, had also just possessed simultaneously to utilize voice to carry out the ability of device control.Utilizing voice to carry out control apparatus can understand to increase by following explanation.
Program is under waiting status, receive the input of voice, when phonetic entry, the voice of reception and the information of the storage in the memory are compared, 4000 begin in the address of 47FF the content correspondence order, 4800 to 4FFF is another order, the voice of input are through amplifying, after treatment circuit 3 through having multi-channel a/d converter converts digital quantity to, so digital quantity is close with the digital quantity of 4000 to 4FFF each section quantifications, the treatment circuit 3 that has multi-channel a/d converter will be controlled the I/O port action of a correspondence, as the corresponding voice command 1 of P1.0, corresponding voice command 2 of P1.1 or the like.
The wave filter of Cai Yonging can be the anti aliasing distortion filter in the present invention; To avoid sampled signal generation aliasing, sample then and quantize, and use the bank of filters formed by eight wave filters with signal decomposition to different frequency bands.Before carrying out feature extraction, signal also needs through preaccentuator and end-point detection, and its purpose is to remove the noise effect that individual pronunciation difference and environment cause etc.; Then, the parameters,acoustic that reflects the voice signal essential characteristic is extracted, the characteristic parameter that adopts among the present invention mainly is the short-time average amplitude; The time series of the characteristic parameter of voice has just constituted the pattern of voice, and itself and the reference model that has been stored in the machine are compared (pattern match) one by one, and the reference model of being determined to draw optimum matching by decision rule is a recognition result; Reference model obtains before system uses and storage is got up.For this reason, import a series of known voice signals, the feature of extracting them is as the reference pattern, and this is with the procedure initialization process.
Claims (9)
1, voice control method, initialization and two processes of voice control output of comprising voice messaging: the initialization of (1) voice messaging, the initialization of voice messaging is that the voice through confirming are transformed into electric signal through microphone, described electric signal is after amplifying circuit amplifies, obtain the electrical information of same audio-frequency information different frequency bands again through the filtering circuit filtering of a plurality of different frequency bands, the electrical information of these different frequency bands changes into corresponding number of bands amount information storage at a certain address field of memory by the processing circuit processes that has multi-channel a/d converter, and described affirmation process is to finish after being confirmed by the acknowledgement key on the keyboard; Memory can deposit the input of a plurality of different phonetic in, and deposits it in different memory address section, and these different memory address sections have determined different control circuit output;
(2), voice control output, after the initialization acknowledgement key is finished, enter voice control output, after initialization finished in the voice of confirming, so long as import once more through the voice of confirming, treatment circuit compares with digital quantity in having memory digitized phonetic entry by different addresses, when unanimity, corresponding output terminal is moved.
2, the phonetic controller of sound control method design according to claim 1 is characterized in that: which comprises at least: microphone: be used to receive phonetic entry, and change voice input signal into electric signal;
Filtering circuit: the audio-frequency information that is used for a road above-mentioned electric signal is produced simultaneously the output different frequency bands;
The treatment circuit of one band A/D converter is electrically connected with the output of filtering circuit, is used for different frequency bands output changing into the digital quantity storage with filtering circuit;
The digital quantity that is used for and has stored compares, to determine the validity of phonetic entry, and when the voice messaging that reads is effective, the output terminal action of a correspondence of control;
Memory is connected electrically on the data bus of control circuit, is used to store effective speech input information;
Keyboard: be connected electrically in the I/O mouth of control circuit, comprise that at least a button is used to control phonetic entry and whether deposits in the memory; At least also comprise that a button is used for determining that an output port of control circuit is corresponding with a speech input information, when the phonetic entry of correspondence, control circuit moves corresponding output end.
3, phonetic controller according to claim 2 is characterized in that: the control circuit 4 that has multi-channel a/d converter can adopt the AT89S8535 integrated circuit.
4, phonetic controller according to claim 2 is characterized in that: the control circuit 4 that has multi-channel a/d converter also can be selected the integrated circuit of other band A/D converter;
5, phonetic controller according to claim 2 is characterized in that: multichannel filtering circuit 3 can be chosen 3 to 8 audio sections;
6, phonetic controller according to claim 2 is characterized in that: be electrically connected with electronic switch 7 on the I/O mouth, electronic switch 7 can replace the button of existing household electrical appliance.
7, phonetic controller according to claim 2 is characterized in that: the wave filter of employing can be the anti aliasing distortion filter
8, phonetic controller according to claim 2 is characterized in that: before carrying out feature extraction, signal also needs through preaccentuator and end-point detection, and its purpose is to remove the noise effect that individual pronunciation difference and environment cause etc.; Then, the parameters,acoustic with reflection voice signal essential characteristic extracts.
9, phonetic controller according to claim 2 is characterized in that: described characteristic parameter mainly is the short-time average amplitude; The time series of the characteristic parameter of voice has just constituted the pattern of voice, and itself and the reference model that has been stored in the machine are compared (pattern match) one by one, and the reference model of being determined to draw optimum matching by decision rule is a recognition result; Reference model obtains before system uses and storage is got up.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN02114534A CN1452159A (en) | 2002-04-18 | 2002-04-18 | Speech controlling device and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN02114534A CN1452159A (en) | 2002-04-18 | 2002-04-18 | Speech controlling device and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1452159A true CN1452159A (en) | 2003-10-29 |
Family
ID=29220933
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN02114534A Pending CN1452159A (en) | 2002-04-18 | 2002-04-18 | Speech controlling device and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1452159A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1779782B (en) * | 2004-11-26 | 2010-04-21 | 佳能株式会社 | User interface design apparatus and method |
| CN101165779B (en) * | 2006-10-20 | 2010-06-02 | 索尼株式会社 | Information processing apparatus and method, program, and record medium |
| CN101833292A (en) * | 2010-04-30 | 2010-09-15 | 中山大学 | Digital home control method, controller and system |
| CN104702697A (en) * | 2015-03-24 | 2015-06-10 | 蒋睿 | Intelligent tone tuning system based on Internet of things |
| CN104766462A (en) * | 2014-12-29 | 2015-07-08 | 深圳市欧润特科技有限公司 | Sound wave remote control system and sound wave remote control method |
| CN105225663A (en) * | 2015-08-26 | 2016-01-06 | 成都视达科信息技术有限公司 | A kind of method and system utilizing acoustic control to carry out multi-screen interactive equipment to be located by connecting |
| CN106054583A (en) * | 2016-07-19 | 2016-10-26 | 柳州六品科技有限公司 | Bracelet type wireless alarm wake-up device |
-
2002
- 2002-04-18 CN CN02114534A patent/CN1452159A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1779782B (en) * | 2004-11-26 | 2010-04-21 | 佳能株式会社 | User interface design apparatus and method |
| CN101165779B (en) * | 2006-10-20 | 2010-06-02 | 索尼株式会社 | Information processing apparatus and method, program, and record medium |
| CN101833292A (en) * | 2010-04-30 | 2010-09-15 | 中山大学 | Digital home control method, controller and system |
| CN104766462A (en) * | 2014-12-29 | 2015-07-08 | 深圳市欧润特科技有限公司 | Sound wave remote control system and sound wave remote control method |
| CN104702697A (en) * | 2015-03-24 | 2015-06-10 | 蒋睿 | Intelligent tone tuning system based on Internet of things |
| CN105225663A (en) * | 2015-08-26 | 2016-01-06 | 成都视达科信息技术有限公司 | A kind of method and system utilizing acoustic control to carry out multi-screen interactive equipment to be located by connecting |
| CN106054583A (en) * | 2016-07-19 | 2016-10-26 | 柳州六品科技有限公司 | Bracelet type wireless alarm wake-up device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106448654A (en) | Robot speech recognition system and working method thereof | |
| CN106601245A (en) | Vehicle-mounted intelligent audio device and audio processing method | |
| CN1452159A (en) | Speech controlling device and method | |
| CN206312566U (en) | A kind of vehicle intelligent audio devices | |
| CN109389976A (en) | Intelligent household electrical appliance control method and device, intelligent household electrical appliance and storage medium | |
| CN203799645U (en) | Microphone-array-based multichannel voice processing apparatus | |
| CN112017658A (en) | Operation control system based on intelligent human-computer interaction | |
| CN111105798B (en) | Equipment control method based on voice recognition | |
| CN107799118A (en) | Voice directions recognition methods and apparatus and system, home controller | |
| JP2759267B2 (en) | Method and apparatus for synthesizing speech from a speech recognition template | |
| CN1707613A (en) | Collecting apparatus and method for noise insulation audio frequency | |
| CN111192573B (en) | Intelligent control method for equipment based on voice recognition | |
| CN108735230A (en) | Background music recognition methods, device and equipment based on mixed audio | |
| CN204288802U (en) | Voice control electric appliance | |
| CN113516975A (en) | Intelligent household voice-operated switch system and control method | |
| CN2535897Y (en) | Voice control device | |
| CN208586436U (en) | A kind of intelligent sound washing machine | |
| CN210575088U (en) | Voice recognition household appliance control device | |
| CN104766610A (en) | Voice recognition system and method based on vibration | |
| CN103731768A (en) | Sound pickup method and device | |
| CN106448676A (en) | Robot speech recognition system based on natural language processing | |
| CN109949831B (en) | Method and device for voice recognition in intelligent equipment and computer readable storage medium | |
| CN100396067C (en) | DTMF (double-tone multifrequency) signal receiving method and apparatus | |
| CN111292748A (en) | Voice input system capable of adapting to various frequencies | |
| CN215221174U (en) | Socket capable of controlling clothes airing machine through voice |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |