CN102546034A - Method and equipment for processing voice signals - Google Patents
Method and equipment for processing voice signals Download PDFInfo
- Publication number
- CN102546034A CN102546034A CN201210026441XA CN201210026441A CN102546034A CN 102546034 A CN102546034 A CN 102546034A CN 201210026441X A CN201210026441X A CN 201210026441XA CN 201210026441 A CN201210026441 A CN 201210026441A CN 102546034 A CN102546034 A CN 102546034A
- Authority
- CN
- China
- Prior art keywords
- data
- functionalities
- present group
- group data
- module
- 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.)
- Granted
Links
Images
Abstract
The embodiment of the invention discloses a method and equipment for processing voice signals. The method comprises the following steps of: carrying out quantization encoding on acquired inputted current voice signals to obtain current group data; searching from a set formed by functional group data corresponding to voice control functions to obtain the functional group data matched with the current group data; and executing a voice control function corresponding to the functional group data obtained by searching. Users can achieve the voice control over digital equipment, can more conveniently operate and do not worry about that the variety and the poor quality of a remote controller any more, the manufacture cost of the digital equipment is lowered, the remote controller is dispensed with being configured, and further, convenience is brought for handicapped people.
Description
Technical field
The present invention relates to the signal processing field, relate in particular to a kind of audio signal processing method and equipment.
Background technology
A kind of as digital device, STB all will carry out man-machine interaction in the process of Development and Production now, and interactive means generally has two kinds, and the first adopts remote controller to carry out straighforward operation, and it two is to carry out button operation through front panel.
Though remote controller is portable, along with the development of society, household electrical appliance are more and more, and remote controller is also more and more.Consequently obscure easily and lose, remote controller also than damaging easily or malfunctioning, had both increased cost and had let the client use inconvenience.The front panel user operates just more inconvenient.For can't manually operated disabled person, the operation of remote controller or front panel is inconvenience more especially.
Summary of the invention
Embodiment of the invention technical problem to be solved is, a kind of audio signal processing method and equipment are provided, and carries out the control operation of digital device to make things convenient for the user.
In order to solve the problems of the technologies described above, the embodiment of the invention has proposed a kind of audio signal processing method, comprising:
Gather the current speech signal of input;
Said current speech signal is carried out quantization encoding obtain the present group data;
From form set by the functionalities data corresponding, search the functionalities data that obtain with said present group Data Matching with voice control function;
Carry out and the said corresponding voice control function of functionalities data that obtains of searching.
Correspondingly, the embodiment of the invention also provides a kind of voice signal treatment facility, comprising:
Acquisition module is used to gather the current speech signal of input;
Coding module is used for that said current speech signal is carried out quantization encoding and obtains the present group data;
DBM, the functionalities data that are used to store by corresponding with voice control function form set;
Search module, be used for, search the functionalities data that obtain with said present group Data Matching from said set;
Executive Module is used to carry out and the said corresponding voice control function of functionalities data that obtains of searching.
The embodiment of the invention is carried out quantization encoding to the current speech signal of gathering input and is obtained the present group data through a kind of audio signal processing method and equipment are provided, secondly from formed set by the functionalities data corresponding with voice control function; Search the functionalities data that obtain with said present group Data Matching; Carry out then and search the corresponding voice control function of functionalities data that obtains, thereby the user can realize the voice control to digital device, makes the user more convenient to operate with said; No longer be that remote controller is many and bad anxious; Reduced the manufacturing cost of digital device, can join remote controller, more the disabled person provides convenience.
Description of drawings
Fig. 1 is the main flow chart of the audio signal processing method of the embodiment of the invention.
Fig. 2 is the assignment procedure of the functionalities data of the embodiment of the invention.
Fig. 3 is the sub-process figure of 103 steps in the audio signal processing method of the embodiment of the invention.
Fig. 4 is the sub-process figure of 101 steps in the audio signal processing method of the embodiment of the invention.
Fig. 5 is the preemphasis process chart in the audio signal processing method of the embodiment of the invention.
Fig. 6 is the flow chart of assignment procedure of the functionalities data of the embodiment of the invention.
Fig. 7 is a kind of application diagrammatic sketch of the audio signal processing method of the embodiment of the invention.
Fig. 8 is the schematic flow sheet of the another kind of audio signal processing method of the embodiment of the invention.
Fig. 9 is the primary structure figure of the voice signal treatment facility of the embodiment of the invention.
Figure 10 is the structure chart of searching module 904 of the voice signal treatment facility of the embodiment of the invention.
Figure 11 is the element assembly drawing of the voice signal treatment facility of the embodiment of the invention.
Embodiment
Below in conjunction with accompanying drawing, the embodiment of the invention is elaborated.
With reference to Fig. 1, the audio signal processing method of the embodiment of the invention mainly comprises following flow process:
101, gather the current speech signal of input;
102, the current speech signal is carried out quantization encoding obtain the present group data;
103, from form set by the functionalities data corresponding, search the functionalities data that obtain with the present group Data Matching with voice control function;
104, carry out and search the corresponding voice control function of functionalities data that obtains.
Particularly, after the user imports the current speech signal through microphone, through it is carried out quantization encoding; And carry out the coupling of functionalities data, thus the functionalities data of coupling gained are carried out corresponding voice control function, realized voice control to digital device; Making the user more convenient to operate, no longer is that remote controller is many and bad anxious, has reduced the manufacturing cost of digital device; Can join remote controller, more the disabled person provides convenience.
As a kind of execution mode, said method also comprises the assignment procedure of functionalities data as shown in Figure 2, and assignment procedure and 101-102 are similar:
201, gather the function voice signal of input;
202, said function voice signal is carried out quantization encoding, obtain said functionalities data.
Particularly, the user is through behind the microphone input function voice signal, and through it is carried out quantization encoding, it is used for searching of flow process shown in Figure 1 to obtain the functionalities data.
As a kind of execution mode, 103 can specifically comprise flow process as shown in Figure 3:
301, confirm present group data A and alternative functionalities data B editing distance LD (A, B), and the longest common subset data length LCS of present group data A and functionalities data B (A, B);
302, according to editing distance LD (A, B) with the longest common subset data length LCS (A, ratio B), confirm present group data A and alternative functionalities data B degree of approximation S (A, B), degree of approximation S (A, B) can be by definite as shown in the formula (1):
303, (A B), searches the functionalities data B that obtains with present group data A coupling according to degree of approximation S.
Particularly, when the degree of approximation of tolerance present group data A and alternative functionalities data B, more rational similarity should satisfy following character:
One: 0≤S of character (A, B)≤100%, 0 expression is dissimilar fully, and 100% expression is equal fully,
Character two: S (A, B)=S (B, A).
Adopt following formula (1) can satisfy character one and character two exactly, for example:
Present group data A=GGATCGA, B=GAATTCAGTTA, through calculating, LD (A, B)=5, LCS (A, B)=2, present group data A and alternative functionalities data B mate as shown in table 1 below:
| ?A： | ?G | ?G | A | ?_ | ?T | ?C | ?_ | ?G | _ | _ | A |
| ?B： | ?G | ?A | A | T | ?T | ?C | A | ?G | ?T | T | A |
Table 1
LD in the formula (1) (A, B)+(this is unique to LCS for A, the B) length of the optimum Match group data of expression present group data A and alternative functionalities data B.It is also noted that in addition, and LD (A, B)+(A is B) with Max (Len (A), Len (B)) and not exclusively equal for LCS.
Editing distance LD (A; That B) when calculating, adopt is editing distance algorithm (Edit Distance; ED), it is to represent the difference between present group data A and the alternative functionalities data B with the number of operations that present group data A becomes alternative functionalities data B through insertion, deletion, replacement data.Its algorithm can be as following:
The longest common subset data length LCS (A, computational algorithm B) can be as following:
As a kind of execution mode, 101 can comprise flow process as shown in Figure 4:
401, the current speech signal is sampled with the frequency of 9kbit/s;
402, adopt 12 A/D converters to carry out mould/number conversion to the first order signal of sampling;
403, utilize firstorder filter to carry out preemphasis to the second level signal after mould/number conversion and handle.
Particularly, above-mentioned 101 is the Signal Pretreatment part, processing such as the current speech signals sampling of main completion input, mould/number conversion.Sample frequency is 9kbit/s, and mould/number conversion is realized by embedded 12 A/D converters.Because the HFS energy of voice signal frequency spectrum is less; Its amplitude is less, and the influence that is interfered was easily being carried out before 102; Can strengthen its HFS earlier; Promptly adopt firstorder filter (1-az-1) to utilize software to carry out preemphasis, wherein a desirable 0.95 or other numerical value, process can be as shown in Figure 5.
Obtain the present group data and can be and adopt the adaptive difference pulse code modulation (Adaptive Differential Pulse-Code Modulation ADPCM) carries out quantization encoding to the current speech signal and the current speech signal is carried out quantization encoding.Particularly, in GSM standard, regular code adopts APCM that selected regular pulses sequence is carried out quantization encoding, and the regular pulses sequence has 13 sampling point X
m(i) form, find earlier | X
m(i) | the X of middle maximum
Max, to X
MaxObtain with 6bit logarithm quantization encoding, to X
MaxThe X ' that obtains after the decoding
MaxBe used for that 13 non-zero sample values are done the normalization processing and obtain X ' (i), that is:
X′(i)＝X′
m(i)/X′
max，i＝0，1，2，...，12
Then, with 3bit uniform quantization X ' (i), every frame needs the 180bit pulse sequence to encode.Because the figure place of each sample signal is too big; When directly the amplitude of input signal being carried out quantization encoding, it is more to consume cpu resource, add digital terminal equipment such as STB just store with data more right; Need not recover voice, so voice quality is not very important.Adopt the ADPCM mode to improve in the embodiment of the invention; Difference to the regular pulses of the regular pulses of reality and prediction is encoded; Each non-zero sample value is only encoded with 2bit; And every frame is encoded with the 104bit pulse sequence, makes code rate reduce to 9kbit/s from 13kbit/s, has also reduced the consumption resource quantity simultaneously.
When using, the assignment procedure of above-mentioned functions group data can comprise process as shown in Figure 6:
601, set top box front panel selects the user to record, and wherein, number of users is configurable;
602, detect the phonetic entry of " I am XXX ";
603, carry out data extract and GSM coding;
604, GSM coding and storage audio user database;
605, judge whether times of collection is 3 times, if, then carry out 604, otherwise, carry out 602;
606, front panel is selected that function is set and is recorded;
607, the measuring ability phonetic entry;
608, GSM coding and memory function audio database;
609, judge whether times of collection is 3 times, if, then carry out 610, otherwise, carry out 607;
610, whether arbitration functions is recorded and is accomplished, if, then finish, otherwise, carry out 606.
When using, the audio signal processing method of above-mentioned 101-104 can comprise process as shown in Figure 7:
701, whether detection user " I am XXX " logins effective, if then carry out 702, otherwise carry out 701 again;
702, the measuring ability phonetic entry;
703, data extract and GSM coding;
704 and database in recorded user speech and mated, if mate successfully, then carry out 705, otherwise, execution 702;
705, the function voice response is handled, and returns and carry out 701.
In order to improve searching accuracy, there are at least two functionalities data in each voice control function correspondence, like this; Have a plurality of functionalities data to supply to carry out matched and searched with the present group data, accuracy rate can improve, but can make device responds slack-off simultaneously; As shown in Figure 8, for example:
801, the degree of approximation value S of judgement present group data and 3 functionalities data (A, B) whether all greater than 70, if, then relatively success, otherwise, carry out 802;
802, the degree of approximation value S of judgement present group data and 3 functionalities data (A, B) whether all less than 30, if, then relatively failure, otherwise, carry out 803;
803, judge whether to exist in present group data and 3 the functionalities data one of them degree of approximation value S (A, B) greater than 70, if, then relatively success, otherwise, relatively failure.
Correspondingly, the voice signal treatment facility of the embodiment of the invention mainly comprises structure as shown in Figure 9:
Particularly, search module 904 and can comprise structure shown in figure 10:
Pretreatment module 1001 is used for confirming the editing distance of present group data and functionalities data and the longest common subset data length of present group data and functionalities data;
Degree of approximation computing module 1002 is used for confirming the degree of approximation of present group data and functionalities data according to the ratio of editing distance with the longest common subset data length;
Matching module 1003 is used for according to the degree of approximation, searches the functionalities data that obtain with the present group Data Matching.
Its element assembling of the voice signal treatment facility of the invention described above embodiment can be shown in figure 11, and it mainly includes the LM386 as the core processor controls, and other capacitance resistance elements.
Need to prove that above-mentioned voice signal treatment facility can be STB, DTV, computer or other digital terminal equipments.
Above-mentioned audio signal processing method and equipment can adopt following operational means when using:
At first, need before the user recording to select Customs Assigned Number, record the user name " I am XXX " of oneself, number of users is 4 or can decides with the top-set hardware configuration;
Secondly, get into STB OSD through the front panel operation and be provided with, pronunciation is recorded to common function successively, and sound-recording function mainly is to be main according to button, almost is with a mouthful replacement remote controller, not necessarily wants standard mandarin.Common function recording have (login user name, 0-9, affirmation, cancellation, up and down, withdraw from, standby), other function recording have (menu, information, EPG, record, time shift, F.F., rewind down, stop, previous, next, TV/RADIO, recreation etc.);
Then, each recording is gathered 3 times, and the times of collection user is provided with (it is fixed to look the top-set hardware configuration), and times of collection is many more, and accuracy rate can be high more, but the STB response can be slack-off;
Afterwards, after " I am XXX " acoustic control was landed, the user just can acoustic control, and STB detects the voice data of login user every, and the coupling audio database makes corresponding the processing after the success.
It is more convenient that this acoustic control intelligent terminal system characteristic can make the user use, and no longer is that remote controller is many and bad anxious, and product cost reduces can join remote controller, and more the disabled person provides convenience.
In addition; One of ordinary skill in the art will appreciate that all or part of flow process that realizes in the foregoing description method; Be to instruct relevant hardware to accomplish through program; Described program can be stored in the computer read/write memory medium, and this program can comprise the flow process like the embodiment of above-mentioned each side method when carrying out.Wherein, described storage medium can be magnetic disc, CD, read-only storage memory body (Read-Only Memory, ROM) or at random store memory body (Radom Access Memory, RAM) etc.
The above is an embodiment of the present invention; Should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; Can also make some improvement and retouching, these improvement and retouching also are regarded as protection scope of the present invention.
Claims (10)
1. an audio signal processing method is characterized in that, comprising:
Gather the current speech signal of input;
Said current speech signal is carried out quantization encoding obtain the present group data;
From form set by the functionalities data corresponding, search the functionalities data that obtain with said present group Data Matching with voice control function;
Carry out and the said corresponding voice control function of functionalities data that obtains of searching.
2. the method for claim 1 is characterized in that, said method also comprises the assignment procedure of said functionalities data:
Gather the function voice signal of input;
Said function voice signal is carried out quantization encoding, obtain said functionalities data.
3. the method for claim 1 is characterized in that, at least two said functionalities data of the corresponding existence of each said voice control function.
4. the method for claim 1 is characterized in that, searches the functionalities data that obtain with said present group Data Matching and comprises:
Confirm the editing distance of said present group data and functionalities data, and the longest common subset data length of said present group data and functionalities data;
According to the ratio of said editing distance, confirm the degree of approximation of said present group data and functionalities data with the longest common subset data length;
According to the said degree of approximation, search the functionalities data that obtain with said present group Data Matching.
5. method as claimed in claim 4 is characterized in that, the said degree of approximation is confirmed by following method:
Wherein, A is said present group data, and B is said functionalities data; S (A; B) be the degree of approximation of said present group data and functionalities data, (A B) is the editing distance of said present group data and functionalities data to LD; (A B) is the longest common subset data length of said present group data and functionalities data to LCS.
6. the method for claim 1 is characterized in that, the current speech signal of gathering input comprises:
Said current speech signal is sampled with the frequency of 9kbit/s;
First order signal to said sampling adopts 12 A/D converters to carry out mould/number conversion;
Utilize firstorder filter to carry out preemphasis to the second level signal after said mould/number conversion and handle,
Said current speech signal is carried out quantization encoding to be obtained the present group data and is specially:
Adopt the adaptive difference pulse code modulation that said current speech signal is carried out quantization encoding.
7. a voice signal treatment facility is characterized in that, comprising:
Acquisition module is used to gather the current speech signal of input;
Coding module is used for that said current speech signal is carried out quantization encoding and obtains the present group data;
DBM, the functionalities data that are used to store by corresponding with voice control function form set;
Search module, be used for, search the functionalities data that obtain with said present group Data Matching from said set;
Executive Module is used to carry out and the said corresponding voice control function of functionalities data that obtains of searching.
8. equipment as claimed in claim 7 is characterized in that, said equipment also comprises:
Presetting module is used for after the function voice signal of the input that said acquisition module is gathered, and said coding module is carried out the said functionalities data that quantization encoding obtains to said function voice signal deposit said DBM in.
9. equipment as claimed in claim 7 is characterized in that, the said module of searching comprises:
Pretreatment module is used for confirming the editing distance of said present group data and functionalities data and the longest common subset data length of said present group data and functionalities data;
Degree of approximation computing module is used for confirming the degree of approximation of said present group data and functionalities data according to the ratio of said editing distance with the longest common subset data length;
Matching module is used for according to the said degree of approximation, searches the functionalities data that obtain with said present group Data Matching.
10. like each described equipment in the claim 7 to 9, it is characterized in that said equipment is STB, DTV or computer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201210026441 CN102546034B (en) | 2012-02-07 | 2012-02-07 | Method and equipment for processing voice signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201210026441 CN102546034B (en) | 2012-02-07 | 2012-02-07 | Method and equipment for processing voice signals |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102546034A true CN102546034A (en) | 2012-07-04 |
| CN102546034B CN102546034B (en) | 2013-12-18 |
Family
ID=46352077
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201210026441 Active CN102546034B (en) | 2012-02-07 | 2012-02-07 | Method and equipment for processing voice signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102546034B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104718007A (en) * | 2012-10-04 | 2015-06-17 | 迪士尼企业公司 | Interactive objects for immersive environment |
| WO2017020211A1 (en) * | 2015-08-02 | 2017-02-09 | 李强生 | Method and remote controller for information reminder when matching voice to household electrical appliance |
| CN106815196A (en) * | 2015-11-27 | 2017-06-09 | 北京国双科技有限公司 | Soft text represents number of times statistical method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1223508A (en) * | 1998-01-13 | 1999-07-21 | 黄金富 | Sound-controlled infrared ray remote controller |
| US7138927B2 (en) * | 2000-03-10 | 2006-11-21 | Fang Calvin C | Universal remote controller with voice and digital memory |
| CN101178897A (en) * | 2007-12-05 | 2008-05-14 | 浙江大学 | Speaking man recognizing method using base frequency envelope to eliminate emotion voice |
| CN101516005A (en) * | 2008-02-23 | 2009-08-26 | 华为技术有限公司 | Speech recognition channel selecting system, method and channel switching device |
| CN102023854A (en) * | 2009-09-18 | 2011-04-20 | 上海智问软件技术有限公司 | Template-based semantic variable extraction method |
-
2012
- 2012-02-07 CN CN 201210026441 patent/CN102546034B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1223508A (en) * | 1998-01-13 | 1999-07-21 | 黄金富 | Sound-controlled infrared ray remote controller |
| US7138927B2 (en) * | 2000-03-10 | 2006-11-21 | Fang Calvin C | Universal remote controller with voice and digital memory |
| CN101178897A (en) * | 2007-12-05 | 2008-05-14 | 浙江大学 | Speaking man recognizing method using base frequency envelope to eliminate emotion voice |
| CN101516005A (en) * | 2008-02-23 | 2009-08-26 | 华为技术有限公司 | Speech recognition channel selecting system, method and channel switching device |
| CN102023854A (en) * | 2009-09-18 | 2011-04-20 | 上海智问软件技术有限公司 | Template-based semantic variable extraction method |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104718007A (en) * | 2012-10-04 | 2015-06-17 | 迪士尼企业公司 | Interactive objects for immersive environment |
| US10067557B2 (en) | 2012-10-04 | 2018-09-04 | Disney Enterprises, Inc. | Interactive objects for immersive environment |
| WO2017020211A1 (en) * | 2015-08-02 | 2017-02-09 | 李强生 | Method and remote controller for information reminder when matching voice to household electrical appliance |
| CN106815196A (en) * | 2015-11-27 | 2017-06-09 | 北京国双科技有限公司 | Soft text represents number of times statistical method and device |
| CN106815196B (en) * | 2015-11-27 | 2020-07-31 | 北京国双科技有限公司 | Soft text display frequency statistical method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102546034B (en) | 2013-12-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101002254B (en) | Device and method for robustry classifying audio signals, method for establishing and operating audio signal database | |
| CN109493881B (en) | Method and device for labeling audio and computing equipment | |
| KR102046728B1 (en) | Method and device for identifying time information from voice information | |
| CN105872838A (en) | Sending method and device of special media effects of real-time videos | |
| CN102568478A (en) | Video play control method and system based on voice recognition | |
| CN105095433A (en) | Recommendation method and device for entities | |
| CN107533850B (en) | Audio content identification method and device | |
| CN102045553A (en) | Multimedia transcoding device and method and multimedia player | |
| CN104123930A (en) | Guttural identification method and device | |
| CN103533391A (en) | Two-way interaction digital television box system with acoustic control type interaction and implementation method | |
| CN103336788A (en) | Humanoid robot added Internet information acquisition method and system | |
| CN103347070B (en) | Push method, terminal, server and the system of speech data | |
| CN104142936A (en) | Audio and video match method and audio and video match device | |
| CN102546034B (en) | Method and equipment for processing voice signals | |
| CN104965594A (en) | Intelligent face identification cloud sound control method, device and system thereof | |
| CN106202501A (en) | A kind of information analysis system | |
| CN105335466A (en) | Audio data retrieval method and apparatus | |
| CN103294696A (en) | Audio and video content retrieval method and system | |
| CN109862408A (en) | A kind of user speech identification control method for smart television voice remote controller | |
| CN108776450B (en) | Floor sweeping robot service system and computer readable storage medium | |
| CN105657146B (en) | A kind of communication information prompt method and device | |
| CN109451254A (en) | A kind of smart television digital receiver | |
| CN111079854A (en) | Information identification method, device and storage medium | |
| CN203658764U (en) | User feedback monitoring device | |
| CN109727602A (en) | A kind of method for recognizing sound-groove and device of mobile device terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C53 | Correction of patent of invention or patent application | ||
| CB03 | Change of inventor or designer information |
Inventor after: Cao Lihua Inventor after: Dong Xuefeng Inventor before: Shu Guoling Inventor before: Cao Lihua Inventor before: Dong Xuefeng |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |