CN105023571A - Voice feature extraction control method for recording pen - Google Patents
Voice feature extraction control method for recording pen Download PDFInfo
- Publication number
- CN105023571A CN105023571A CN201510447252.3A CN201510447252A CN105023571A CN 105023571 A CN105023571 A CN 105023571A CN 201510447252 A CN201510447252 A CN 201510447252A CN 105023571 A CN105023571 A CN 105023571A
- Authority
- CN
- China
- Prior art keywords
- voice signal
- recording pen
- phonetic feature
- voice
- speech
- 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
Landscapes
- User Interface Of Digital Computer (AREA)
Abstract
The invention discloses a voice feature extraction control method for recording pens, which comprises the steps of inputting a language signal; detecting and identifying the input voice signal to obtain the voice characteristics and store the voice characteristics; identifying the input voice signal continuously and accumulating the frequency of the voice characteristics when the stored voice characteristics reappear; judging if the detected voice signal is ended or not and going back to the identification step when the voice signal detection is not ended; executing the extraction step when the voice signal detection is ended; extracting the voice characteristics that has the highest accumulate frequency after the voice signal detection is ended; and taking the extracted voice characteristics as the keywords of the language signal for outputting. According to the invention, not only logging in can be down, automatic detection and identification can be realized; the key words can be extracted automatically; the key words of the voice signal can be obtained without opening the files one by one when a user is checking the audio frequency files; the voice signal processing process is more intelligent; and the user p can identify the audio frequency files conveniently.
Description
Technical field
The present invention relates to a kind of speech feature extraction control method for recording pen, belong to the technical field that recording pen controls.
Background technology
Digital recording pen, also referred to as digital recording clave or digital voice recorder, the one of digital voice recorder, for the ease of operating and promote recording quality moulding not based on simple pen type, easy to carry, have several functions, as function of laser pointer, FM frequency modulation, MP3 play simultaneously.Compared with traditional sound-track engraving apparatus, digital recording pen carrys out record audio by the mode of stored digital.
Its Principle of Process is, digital recording pen by the sampling of simulating signal, encoding is converted to digital signal by simulating signal by digital to analog converter, and to store after carrying out certain compression.Even and if digital signal is through multiple copies, acoustic information also can not incur loss, and keeps intact constant.
Because be sound pick-up outfit, the length of record length is the most important technical indicator of digital recording pen naturally.According to the difference of flash capacity, compression algorithm between different product, the length of record length also has very large difference.The recording storage time inside saving as the digital recording pen of 1G, battery stream time, generally at 2-26 hours, can meet most of human needs all at 20-272 hour.If but it is noted that very long record length is because it obtains by employing high compression rate, often affect the quality of recording.
Therefore, make recording pen record type compressibility very low, tonequality is very good, but volume ratio is comparatively large, and general being applicable to requires that higher occasion uses, such as, for Leading Speaches makes the record etc. of file formula.
Although existing recording pen is in Recording Process, realize its function, it still has problems.As application number: in the file of 201310710630.3 applyings date: 2013-12-20, disclose " a kind of recording pen with Hygrothermograph; it comprises recording pen body; it is characterized in that: the front surface of recording pen body has display screen and function button; there is Hygrothermograph the rear surface of recording pen body, and Hygrothermograph is connected with the output circuit of recording pen body interior power supply ".This recording pen is owing to being integrated on recording pen body by Hygrothermograph, add the outdoor practical of recording pen, can not only pass through recording pen recorded speech travel notes during outdoor travel, can also understand the humiture information in air at any time, having more humiture information can correctly carry corresponding equipment.
And at other one section of application number: in the file of 201520029314.4 applyings date: 2015-01-15, disclose " a kind of two recording pen stored; described recording pen comprises microphone, audio coding decoding module, storage control circuit and the first storage medium, wherein: described microphone is used for the voice signal of reception to be converted into simulated audio signal; Described audio coding decoding module is connected with described microphone, samples and be encoded to digital audio and video signals to the simulated audio signal received; Described storage control circuit is connected with described audio coding decoding module, and the digital audio and video signals based on coding generates recording file, and described recording file is written to described first storage medium and the second storage medium; Described first storage medium, is connected with described storage control circuit, for storing described recording file ", the two recording pen stored of embodiment of this utility model can preserve recording file in two storage mediums simultaneously, improves the security of recording file.
Although above-mentioned document makes improvement to recording pen, make function more perfect.But it is in practice process, still existing defects.Existing recording pen in use, voice input function can only be performed, and keyword extraction cannot be carried out to voice, user needs to listen to recording just can know its content, and recording pen cannot extract frequency appearance phonetic feature repeatedly automatically, do not possess thus and automatically detect recognition function, reduce service efficiency.
Summary of the invention
Technical matters to be solved by this invention is to overcome the deficiencies in the prior art, a kind of speech feature extraction control method for recording pen is provided, solves existing recording pen and do not possess and automatically detect recognition function, automatically cannot extract the problem that phonetic feature repeatedly appears in frequency.
The present invention specifically solves the problems of the technologies described above by the following technical solutions:
For a speech feature extraction control method for recording pen, comprise the following steps:
Step 1, input speech signal;
Step 2, detection and Identification are carried out to inputted voice signal, obtain phonetic feature and also deposit;
Step 3, the continuous voice signal to input identify, when again there is the phonetic feature deposited in voice signal, carrying out frequency add up this phonetic feature;
Step 4, judge detect voice signal whether terminate, at the end of Speech signal detection is not, return step 2; At the end of Speech signal detection, perform step 5;
Step 5, after Speech signal detection terminates, extract frequency and add up the highest phonetic feature;
Step 6, the phonetic feature that step 5 extracted export as the key word of voice signal.
Further, as a preferred technical solution of the present invention: obtain at least two phonetic features in described step 2.
Further, as a preferred technical solution of the present invention: described step 3 also comprises and contrasting accumulative for the frequency of phonetic feature, obtains frequency and adds up the highest phonetic feature.
Further, as a preferred technical solution of the present invention: the key word that extracted phonetic feature generates as voice signal after text exports by described step 5.
Further, as a preferred technical solution of the present invention: described step 5 also comprises the filename extracted phonetic feature being generated voice signal, and is exported with this filename by voice signal.
The present invention adopts technique scheme, can produce following technique effect:
(1) the speech feature extraction control method for recording pen provided by the present invention, by carrying out detection and Identification to the voice signal of input, obtain phonetic feature and add up, at the end of the voice signal detected, extract frequency and add up the highest phonetic feature, and it can be used as the key word of voice signal to export, make voice signal not only can typing thus, can also detection and Identification automatically, automatic extraction key word, make user when checking audio file, without the need to opening file one by one, the key word of voice signal can be known, make Speech processing process more intelligent, facilitate user to the resolution of audio file, the existing recording pen of effective solution does not possess and automatically detects recognition function, automatically cannot extract the problem that phonetic feature repeatedly appears in frequency.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the present invention for the speech feature extraction control method of recording pen.
Embodiment
below in conjunction with Figure of description, embodiments of the present invention are described.
As shown in Figure 1, the invention provides a kind of speech feature extraction control method for recording pen, the method specifically comprises the following steps:
Step 1, input speech signal;
Step 2, detection and Identification are carried out to inputted voice signal, obtain phonetic feature and also deposit;
Step 3, the continuous voice signal to input identify, when again there is the phonetic feature deposited in voice signal, carrying out frequency add up this phonetic feature;
Step 4, judge detect voice signal whether terminate, at the end of Speech signal detection is not, return step 2; At the end of Speech signal detection, perform step 5;
Step 5, after Speech signal detection terminates, extract frequency and add up the highest phonetic feature;
Step 6, the phonetic feature that step 5 extracted export as the key word of voice signal.
Thus, at the end of the voice signal detected, extract frequency and add up the highest phonetic feature, and it can be used as the key word of voice signal to export, make voice signal not only can typing thus, can also detection and Identification automatically, automatically extract key word.
Further, method of the present invention is when implementing, and the phonetic feature at least two obtained in step 2, namely deposits when identifying a phonetic feature, deposits respectively when obtaining another one or multiple phonetic feature; And be at least two identifying can be made sensitiveer.Under multiple phonetic feature mode, after identifying one by one respectively, its frequency values corresponding, occurs then being recorded as 11 time, occur 4 times be then 4, occur being then for N time N.
And, described step 3 can also comprise and contrasting accumulative for the frequency of phonetic feature carrying out process that frequency adds up, the frequency accumulative total 7 obtained as the frequency accumulative total 3 that obtained by phonetic feature A and phonetic feature B contrasts, and it is the highest for obtaining the frequency accumulative total 7 that phonetic feature B obtains; Realize comparison process, make leaching process more accurate,
On this basis, can also improve further in described step 5 in control method of the present invention, the key word that extracted phonetic feature generates as voice signal after text is exported.Namely system often generates an audio file, then automatically generate corresponding text, content of text and key word content, key word is extracted phonetic feature, text can be set to binding pattern with audio file, making user when selecting audio file, automatically text being ejected, being convenient to the main contents that user checks this section audio.
Further, in above process, described step 5 can also comprise the filename extracted phonetic feature being generated voice signal, and is exported with this filename by voice signal; Namely direct using the filename of phonetic feature as this audio file, make user can know the main contents of this audio frequency by rapid interpretive.
To sum up, speech feature extraction control method for recording pen provided by the present invention, make voice signal not only can typing, can also detection and Identification automatically, automatically extracting key word, making user when checking audio file, without the need to opening file one by one, the key word of voice signal can be known, make Speech processing process more intelligent, facilitate user to the resolution of audio file; The existing recording pen of effective solution does not possess and automatically detects recognition function, automatically cannot extract the problem that phonetic feature repeatedly appears in frequency.
By reference to the accompanying drawings embodiments of the present invention are explained in detail above, but the present invention is not limited to above-mentioned embodiment, in the ken that those of ordinary skill in the art possess, can also makes a variety of changes under the prerequisite not departing from present inventive concept.
Claims (5)
1., for a speech feature extraction control method for recording pen, it is characterized in that, comprise the following steps:
Step 1, input speech signal;
Step 2, detection and Identification are carried out to inputted voice signal, obtain phonetic feature and also deposit;
Step 3, the continuous voice signal to input identify, when again there is the phonetic feature deposited in voice signal, carrying out frequency add up this phonetic feature;
Step 4, judge detect voice signal whether terminate, at the end of Speech signal detection is not, return step 2; At the end of Speech signal detection, perform step 5;
Step 5, after Speech signal detection terminates, extract frequency and add up the highest phonetic feature;
Step 6, the phonetic feature that step 5 extracted export as the key word of voice signal.
2., according to claim 1 for the speech feature extraction control method of recording pen, it is characterized in that: in described step 2, obtain at least two phonetic features.
3. according to claim 1 for the speech feature extraction control method of recording pen, it is characterized in that: described step 3 also comprises and contrasting accumulative for the frequency of phonetic feature, obtains frequency and adds up the highest phonetic feature.
4. according to claim 1 for the speech feature extraction control method of recording pen, it is characterized in that: the key word that extracted phonetic feature generates as voice signal after text exports by described step 5.
5. according to claim 1 for the speech feature extraction control method of recording pen, it is characterized in that: described step 5 also comprises the filename extracted phonetic feature being generated voice signal, and is exported with this filename by voice signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510447252.3A CN105023571A (en) | 2015-07-28 | 2015-07-28 | Voice feature extraction control method for recording pen |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510447252.3A CN105023571A (en) | 2015-07-28 | 2015-07-28 | Voice feature extraction control method for recording pen |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105023571A true CN105023571A (en) | 2015-11-04 |
Family
ID=54413490
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510447252.3A Pending CN105023571A (en) | 2015-07-28 | 2015-07-28 | Voice feature extraction control method for recording pen |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105023571A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000341368A (en) * | 1999-05-27 | 2000-12-08 | Nippon Telegr & Teleph Corp <Ntt> | Digital communication equipment, digital communication method and its recording medium |
| CN101894551A (en) * | 2010-07-02 | 2010-11-24 | 华南理工大学 | Method and device for automatically identifying cough |
| CN201781565U (en) * | 2010-08-30 | 2011-03-30 | 康佳集团股份有限公司 | Remote controller, television and television system |
| CN102231278A (en) * | 2011-06-10 | 2011-11-02 | 安徽科大讯飞信息科技股份有限公司 | Method and system for realizing automatic addition of punctuation marks in speech recognition |
| CN102436810A (en) * | 2011-10-26 | 2012-05-02 | 华南理工大学 | Record replay attack detection method and system based on channel mode noise |
| CN103354575A (en) * | 2013-06-14 | 2013-10-16 | 广东欧珀移动通信有限公司 | Method for prompting history conversation content at time of calling or being called, and mobile terminal |
-
2015
- 2015-07-28 CN CN201510447252.3A patent/CN105023571A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000341368A (en) * | 1999-05-27 | 2000-12-08 | Nippon Telegr & Teleph Corp <Ntt> | Digital communication equipment, digital communication method and its recording medium |
| CN101894551A (en) * | 2010-07-02 | 2010-11-24 | 华南理工大学 | Method and device for automatically identifying cough |
| CN201781565U (en) * | 2010-08-30 | 2011-03-30 | 康佳集团股份有限公司 | Remote controller, television and television system |
| CN102231278A (en) * | 2011-06-10 | 2011-11-02 | 安徽科大讯飞信息科技股份有限公司 | Method and system for realizing automatic addition of punctuation marks in speech recognition |
| CN102436810A (en) * | 2011-10-26 | 2012-05-02 | 华南理工大学 | Record replay attack detection method and system based on channel mode noise |
| CN103354575A (en) * | 2013-06-14 | 2013-10-16 | 广东欧珀移动通信有限公司 | Method for prompting history conversation content at time of calling or being called, and mobile terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10977299B2 (en) | Systems and methods for consolidating recorded content | |
| TW202008349A (en) | Speech labeling method and apparatus, and device | |
| CN101002254B (en) | Device and method for robustry classifying audio signals, method for establishing and operating audio signal database | |
| CN103093316B (en) | A kind of bill generation method and device | |
| CN107274916A (en) | The method and device operated based on voiceprint to audio/video file | |
| JP2004229283A (en) | Method for identifying transition of news presenter in news video | |
| CN105245917A (en) | System and method for generating multimedia voice caption | |
| CN103366784B (en) | There is multi-medium play method and the device of Voice command and singing search function | |
| CN102903375A (en) | Music player and play method | |
| EP1536638A4 (en) | Metadata preparing device, preparing method therefor and retrieving device | |
| CN103336773A (en) | System and method for audio and video speech processing and retrieval | |
| CN106373598A (en) | Audio replay control method and apparatus | |
| CN107025913A (en) | A kind of way of recording and terminal | |
| CN104867494A (en) | Naming and classification method and system of sound recording files | |
| CN101825953A (en) | Chinese character input product with combined voice input and Chinese phonetic alphabet input functions | |
| CN102862587B (en) | A kind of railway vehicle machine joint control speech analysis method and equipment | |
| CN106550268B (en) | Video processing method and video processing device | |
| CN111128134B (en) | Acoustic model training method, voice awakening method and device and electronic equipment | |
| CN106528715A (en) | Method and device for checking audio content | |
| CN105070286A (en) | Voice feature extraction control device for recording pen | |
| CN104350545B (en) | Self-recording unit | |
| CN105023571A (en) | Voice feature extraction control method for recording pen | |
| CN205281851U (en) | Electronic reading equipment | |
| CN206313861U (en) | A kind of audio detection system based on the control of NI boards serial ports | |
| KR101382356B1 (en) | Apparatus for forgery detection of audio file |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151104 |
|
| RJ01 | Rejection of invention patent application after publication |