CN105206280A - Information processing method and electronic equipment - Google Patents
Information processing method and electronic equipment Download PDFInfo
- Publication number
- CN105206280A CN105206280A CN201510583838.2A CN201510583838A CN105206280A CN 105206280 A CN105206280 A CN 105206280A CN 201510583838 A CN201510583838 A CN 201510583838A CN 105206280 A CN105206280 A CN 105206280A
- Authority
- CN
- China
- Prior art keywords
- voice data
- audio
- audio frequency
- parameter
- data
- 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
- Circuit For Audible Band Transducer (AREA)
Abstract
The embodiment of the invention provides an information processing method and electronic equipment which are used for correcting sound characteristics of a voice output application program according to needs of a user. The method is applied to the electronic equipment, and the electronic equipment is capable of running a voice frequency output application program. The method includes the steps that based on first operation of the user, first voice frequency data are determined from multiple voice frequency data; first voice frequency parameters are obtained from the first voice frequency data, wherein the first voice frequency parameters are used for representing sound characteristics of the first voice frequency data; second voice frequency data output by the voice frequency output application program are corrected by means of the first voice frequency parameters, so that the second voice frequency data have the sound characteristics.
Description
Technical field
The present invention relates to electronic technology field, particularly relate to a kind of information processing method and electronic equipment.
Background technology
Along with the development of science and technology, electronic technology have also been obtained development at full speed, and the kind of electronic product also gets more and more, and people have also enjoyed the various facilities that development in science and technology brings.Present people can pass through various types of electronic equipment, enjoy the comfortable life along with development in science and technology brings.
Such as, voice output application program has been widely used in the electronic equipments such as smart mobile phone, panel computer or notebook computer, and the life for people brings more facility.Voice output application program can help user search data, start application and managing schedule etc.Meanwhile, voice output application program of the prior art also possesses dialogue function, can engage in the dialogue with user.Such as when user says " good night " voice output application program, voice output application program will be responded user and " wish that you have a pleasant dream ".
Then, voice output application program of the prior art is sounded and mostly is default sound, and therefore user can not select the sound property that voice output application program exports.
Summary of the invention
The embodiment of the present application provides a kind of information processing method and electronic equipment, for according to user need the sound property of voice output application program is corrected.
On the one hand, this application provides a kind of information processing method, be applied to electronic equipment, described electronic equipment can run an audio frequency and export application program, comprising:
Based on first operation of user, from multiple voice data, determine the first voice data;
From described first voice data, obtain the first audio frequency parameter, wherein, described first audio frequency parameter is for characterizing the sound property of described first voice data;
Utilize described first audio frequency parameter to correct the second audio data being exported application program output by described audio frequency, to make described second audio data, there is described sound property.
Optionally, from described first voice data, obtain the first audio frequency parameter, comprising:
The waveform of described first voice data is analyzed, extracts described first audio frequency parameter.
Optionally, utilize described first audio frequency parameter to correct the second audio data being exported application program output by described audio frequency, comprising:
Utilize the first audio frequency parameter to train the nervous system that described audio frequency exports application program, export described second audio data to make described nervous system.
Optionally, utilize the first audio frequency parameter to train the nervous system that described audio frequency exports application program, comprising:
Based on described first audio frequency parameter and described neural current spectral, revise described at least one first parameter neural;
Obtain described nervous system export correction after frequency spectrum;
After judging described correction, whether the similarity of the frequency spectrum of frequency spectrum and described first voice data equals to be greater than first threshold;
When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data equals to be greater than described first threshold, terminate described neural training, and control described nervous system and export described second audio data according to the parameter after correcting;
When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data is less than described first threshold, using the current spectral that frequency spectrum after described correction corrects as next round, perform step: based on described first audio frequency parameter and described neural current spectral, revise described at least one first parameter neural.
Optionally, after the second audio data utilizing described first audio frequency parameter correction to be exported by described output application program, described method also comprises:
Add up the frequency of each dialogue key element in described first voice data;
When the frequency of described dialogue key element exceedes Second Threshold, determine that described dialogue key element is common dialogue key element;
Described common dialogue key element is added in described second audio data.
On the other hand, this application provides a kind of electronic equipment, described electronic equipment can run an audio frequency and export application program, and described electronic equipment comprises:
First determining unit, for the first operation based on user, determines the first voice data from multiple voice data;
First obtains unit, and for obtaining the first audio frequency parameter from described first voice data, wherein, described first audio frequency parameter is for characterizing the sound property of described first voice data;
Correcting unit, for utilizing described first audio frequency parameter to correct the second audio data being exported application program output by described audio frequency, has described sound property to make described second audio data.
Optionally, described first obtains unit is used for analyzing the waveform of described first voice data, extracts described first audio frequency parameter.
Optionally, described correcting unit is trained the nervous system that described audio frequency exports application program for utilizing the first audio frequency parameter, exports described second audio data to make described nervous system.
Optionally, described correcting unit is used for based on described first audio frequency parameter and described neural current spectral, revises described at least one first parameter neural; Obtain described nervous system export correction after frequency spectrum; After judging described correction, whether the similarity of the frequency spectrum of frequency spectrum and described first voice data equals to be greater than first threshold; When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data equals to be greater than described first threshold, terminate described neural training, and control described nervous system and export described second audio data according to the parameter after correcting; When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data is less than described first threshold, using the current spectral that frequency spectrum after described correction corrects as next round, perform step: based on described first audio frequency parameter and described neural current spectral, revise described at least one first parameter neural.
Optionally, described electronic equipment also comprises:
Statistic unit, for adding up the frequency of each dialogue key element in described first voice data;
Second determining unit, for when the frequency of described dialogue key element exceedes Second Threshold, determines that described dialogue key element is common dialogue key element;
Adding device, for adding to described common dialogue key element in described second audio data.
Above-mentioned one or more technical scheme in the embodiment of the present application, at least has one or more technique effects following:
In the technical scheme of the embodiment of the present application, electronic equipment determines based on first operation of user the first voice data that user selects from multiple voice data, and then from the first voice data, obtain the first audio frequency parameter of the sound property for characterizing the first voice data.Utilize the first audio frequency parameter to correct the second audio data that audio frequency written-out program exports, make second audio data have the sound property of the first voice data.So, solve the technical matters of the sound property that user in prior art can not select voice output application program to export, achieve the technique effect needing that the sound property of voice output application program is corrected according to user.For user, the first voice data of the sound property selecting oneself to need from multiple voice data, such as user likes the sound of host A, then select to speak the voice data of people for host A as the first voice data.Thus electronic equipment is corrected the second audio data that audio frequency exports application program output.User just can hear that audio frequency exports application program host A voice output second audio data.
Accompanying drawing explanation
Fig. 1 is the information processing method process flow diagram in the embodiment of the present application;
Fig. 2 is the electronic devices structure schematic diagram in the embodiment of the present application.
Embodiment
The embodiment of the present application provides a kind of information processing method and electronic equipment, for solving the technical matters of the sound property that user in prior art can not select voice output application program to export, achieve the technique effect needing that the sound property of voice output application program is corrected according to user.
In order to solve the problems of the technologies described above, the technical scheme general thought that the application provides is as follows:
In the technical scheme of the embodiment of the present application, electronic equipment determines based on first operation of user the first voice data that user selects from multiple voice data, and then from the first voice data, obtain the first audio frequency parameter of the sound property for characterizing the first voice data.Utilize the first audio frequency parameter to correct the second audio data that audio frequency written-out program exports, make second audio data have the sound property of the first voice data.So, solve the technical matters of the sound property that user in prior art can not select voice output application program to export, achieve the technique effect needing that the sound property of voice output application program is corrected according to user.For user, the first voice data of the sound property selecting oneself to need from multiple voice data, such as user likes the sound of host A, then select to speak the voice data of people for host A as the first voice data.Thus electronic equipment is corrected the second audio data that audio frequency exports application program output.User just can hear that audio frequency exports application program host A voice output second audio data.
Below by accompanying drawing and specific embodiment, technical solution of the present invention is described in detail, the specific features being to be understood that in the embodiment of the present application and embodiment is the detailed description to technical scheme, instead of the restriction to technical scheme, when not conflicting, the technical characteristic in the embodiment of the present application and embodiment can combine mutually.
Term "and/or" herein, being only a kind of incidence relation describing affiliated partner, can there are three kinds of relations in expression, and such as, A and/or B, can represent: individualism A, exists A and B simultaneously, these three kinds of situations of individualism B.In addition, character "/" herein, general expression forward-backward correlation is to the relation liking a kind of "or".
The application's first aspect provides a kind of information processing method, as shown in Figure 1, comprises the steps:
S101: based on first operation of user, determine the first voice data from multiple voice data.
S102: obtain the first audio frequency parameter from described first voice data.
S103: utilize described first audio frequency parameter to correct the second audio data being exported application program output by described audio frequency.
Specifically, in the embodiment of the present application, multiple voice data is stored in electronic equipment.Each voice data is the voice data comprising speaker's sound.For example, user can by the voice input device of electronic equipment, and as microphone records one section of audio frequency of oneself speaking, so recorded a voice data in other words person of good sense is the voice data of user oneself.Can download the voice data of other speakers from server, the application does not do concrete restriction yet.
Suppose that multiple voice data is specially 3 voice datas, 1st voice data is illustrate that people is the voice data of user oneself, the voice data of the 2nd voice data to be speaker be L surname female star, the 3rd voice data is illustrate that people is the voice data of Z surname matinée idol.
User operates according to the execution first that needs of oneself, determines the first voice data to make electronic equipment.Specifically, the touch-control clicking operation that the first operational example is carried out as the region for showing the first voice data at the touching display screen of electronic equipment, or select identifier to the first voice data for being moved by directionkeys, and click the operation confirming button.The application those of ordinary skill in the field can be arranged according to actual, and the application does not do concrete restriction.
For convenience of subsequent descriptions, suppose that multiple voice data is specially 3 voice datas, 1st voice data is illustrate that people is the voice data of user oneself, and the voice data of the 2nd voice data to be speaker be L surname female star, the 3rd voice data is illustrate that people is the voice data of Z surname matinée idol.Based on first operation of user, determine that the first voice data is the 2nd voice data.
After determining the first voice data, perform S102: from described first voice data, obtain the first audio frequency parameter.
Specifically, the first audio frequency parameter is for characterizing the sound property of the first voice data.In the embodiment of the present application, sound property comprises tone color and the tone of speaker's sound in the first voice data.First audio frequency parameter includes but not limited to that fundamental frequency, harmonic frequency, harmonic attenuation speed, first-harmonic peak value, harmonic spike, peak value sequence etc. characterize the parameter of tone color and tone, the application those of ordinary skill in the field can be arranged according to actual, and the application does not do concrete restriction.
In S102, when having comprised the first audio frequency parameter in the first voice data, such as the first voice data has specifically comprised audio file and Parameter File, then can read Parameter File, and then obtains the first audio frequency parameter.
In addition, electronic equipment needs to obtain the first audio frequency parameter from the first voice data, can also comprise:
The waveform of described first voice data is analyzed, extracts described first audio frequency parameter.
Specifically, when not comprising the first audio frequency parameter in the first voice data, electronic equipment needs from the first voice data, to extract the first audio frequency parameter voluntarily.
Specifically, in order to obtain the first audio frequency parameter, first electronic equipment obtains the waveform of the first voice data.In the embodiment of the present application, the waveform of the first voice data is specially the first voice data waveform in the time domain.Then, the waveform of the first voice data is analyzed.Such as Fourier transform is carried out to the waveform of the first voice data, thus obtain the frequency spectrum of the first voice data.The peak value sequence etc. of the first-harmonic, first harmonic, second harmonic etc. of the first voice data and the peak value sequence of first-harmonic and harmonic wave can be extracted from the frequency spectrum of the first voice data.The multiple parameters identified the most at last are as the first audio frequency parameter.
Next, after obtaining the first audio frequency parameter, just can export to audio frequency the second audio data that application program exports and correct, thus make second audio data have the sound property of the first voice data.
Optionally, utilize the first voice data to carry out timing to second audio data, the first audio frequency parameter can be utilized to train the nervous system that audio frequency exports application program, export second audio data to make nervous system.
Specifically, in the embodiment of the present application, the audio output function that audio frequency exports application program is realized by nervous system, namely when audio frequency output application program needs outputting audio data, the primary data that audio frequency exports the voice data that application program will export inputs from neural input end, such as input characters " wishes that you have a pleasant dream ", and then nervous system " wishes that you have a pleasant dream " from output terminal output audio.
Therefore, when needing audio frequency output application program outputs to have the second audio data of the first voice data sound property, specifically according to the first audio frequency parameter, nervous system is trained exactly, and then make the audio frequency parameter of the nervous system outputting audio data after training identical with the first audio frequency parameter, or approximate identical.
In the embodiment of the present application, the audio frequency parameter of the nervous system outputting audio data after training and the first audio frequency parameter approximately uniform approximate extents behaviour ear cannot the scopes of identification.Such as the resolution of people's ear to frequency is about 0.1Hz, and namely people's ear hears that two frequency phase-differences are less than or equal to the audio frequency of 0.1Hz, will think that two audio tones are identical.Therefore, when the frequency of nervous system outputting audio data after training and the frequency phase-difference of the first voice data are less than or equal to 0.1Hz, electronic equipment can think that the audio frequency parameter of the nervous system outputting audio data after training is similar to identical with the frequency of the first voice data.
In the embodiment of the present application, utilize the first audio frequency parameter to train nervous system, specifically comprise:
Based on described first audio frequency parameter and described neural current spectral, revise described at least one first parameter neural;
Obtain described nervous system export correction after frequency spectrum;
After judging described correction, whether the similarity of the frequency spectrum of frequency spectrum and described first voice data equals to be greater than first threshold;
When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data equals to be greater than described first threshold, complete described neural training, and control described nervous system and export described second audio data according to the parameter after correcting;
When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data is less than described first threshold, using the current spectral that frequency spectrum after described correction corrects as next round, perform step: based on described first audio frequency parameter and described neural current spectral, revise described at least one first parameter neural.
Specifically, first, in first round training, electronic equipment neuralward system inputs the primary data of the first voice data.In the embodiment of the present application, primary data is the data such as word, code that can input neural expression audio data content, and such as voice data is audio frequency " 123456 ", and so the primary data of voice data is word " 123456 ".
Then, each cynapse in nervous system processes according to the primary data of parameter current to the first voice data, finally exports current audio data.And then electronic equipment obtains the initial spectrum of current audio data.In the embodiment of the present application, initial spectrum also can as frequency spectrum after the correction of first round training, and to be specially the amendment difference of at least one the first parameter of nervous system be 0.
Then, the frequency spectrum of contrast initial spectrum and the first voice data, obtains the frequency spectrum similarity of initial spectrum and the first voice data.Due at least one first parameter unmodified when the first round trains, therefore generally, similarity is less than first threshold.First threshold is such as 99%, 98.9% etc.First threshold is larger, and the sound property of final second audio data is more close to the sound property of the first voice data.
Further, electronic equipment is modified at least one first parameter neural.In specific implementation process, the application those of ordinary skill in the field can conventionally modify at least one first parameter to the mode of cynapse parameter modification, and just it is no longer repeated here.
After revising at least one first parameter, start second and take turns training.Take turns in training second, neuralward system inputs the primary data of the first voice data again, then obtain second from neural output terminal and take turns the current audio data after training, and further based on frequency spectrum after the correction of the second current audio data acquisition correspondence of taking turns.
Similarly, the frequency spectrum of contrast corrected spectrum and the first voice data, obtains the similarity of two frequency spectrums.Further, judge whether the similarity of the frequency spectrum correcting rear frequency spectrum and the first voice data equals to be greater than first threshold.
When similarity equals to be greater than first threshold, show that nervous system has trained the degree of the voice data that can export the first voice data sound property, so terminate neural training.And the primary data of second audio data is inputted at neural input end, make nervous system export the second audio data with the first voice data sound property.
And when similarity is less than first threshold, show that nervous system has not yet trained the degree of the voice data that can export the first voice data sound property, still need to continue training.The current spectral that after electronic equipment takes turns the correction of training using second, frequency spectrum is trained as third round, and again revise at least one first parameter, start third round training.
As similar in said process to neural follow-up training, therefore just it is no longer repeated here.
Continue to use example above, suppose that the primary data of second audio data is for " wishing that you have a pleasant dream ", the nervous system of so now having trained exports second audio data according to the primary data of second audio data, and user will hear that L surname female star is saying oneself " wishing the pleasant dream that you do " seemingly.
Optionally, after second audio data is corrected, can also comprise:
Add up the frequency of each dialogue key element in described first voice data;
When the frequency of described dialogue key element exceedes Second Threshold, determine that described dialogue key element is common dialogue key element;
Described common dialogue key element is added in described second audio data.
Specifically, in the embodiment of the present application, talk with key element and include but not limited to word, Chinese idiom, short sentence and interjection etc.Electronic equipment is added up each dialogue key element that the first voice data occurs, obtains the frequency of each dialogue key element.Then, judge whether the frequency of talking with key element exceedes Second Threshold.When talking with key element and exceeding Second Threshold, show that the dialogue key element that the frequency exceedes Second Threshold often occurs in the first voice data, and then show that the speaker that the first voice data is corresponding is accustomed to often saying that the frequency exceedes the dialogue key element of Second Threshold.So the dialogue key element frequency being exceeded Second Threshold is defined as common dialogue key element.Second Threshold is such as 20,16 etc.Finally, common dialogue key element is added in second audio data.The position of adding common dialogue key element can for optional position, and can be the sentence of each sentence or end of the sentence etc., the application do concrete restriction yet.
Because second audio data has been provided with the sound property of the first voice data, so add common dialogue key element in second audio data after, user can be made not only to hear seemingly, and the speaker of the first voice data is saying second audio data.And the common dialogue key element in second audio data, user will be made more to feel, and speaker is speaking, and enhances Consumer's Experience.
For example, example is above continued to use.Suppose in the 1st voice data each dialogue key element add up, count dialogue key element " good " the frequency be 31 times, the frequency of " quite right " is 3 times, " Ow " the frequency be 2 times, all the other dialogue key elements frequencys be 1 time.Second Threshold is 20.Therefore in each dialogue key element, only have the frequency of " good " to exceed Second Threshold, and then determine that " good " is for common dialogue key element.
Further, in the primary data of second audio data " wishes that you have a pleasant dream ", add common dialogue key element " good ", and be specially and be added on beginning of the sentence, so the primary data of second audio data is just adjusted to " good, to wish that you have a pleasant dream ".Finally, the second audio data after neuralward system input adjustment, exports application program output audio " good, to wish that you have a pleasant dream " to make audio frequency.
Based on the inventive concept same with information processing method in previous embodiment, the application's second aspect also provides a kind of electronic equipment, as shown in Figure 2, comprising:
First determining unit 201, for the first operation based on user, determines the first voice data from multiple voice data;
First obtains unit 202, and for obtaining the first audio frequency parameter from described first voice data, wherein, described first audio frequency parameter is for characterizing the sound property of described first voice data;
Correcting unit 203, for utilizing described first audio frequency parameter to correct the second audio data being exported application program output by described audio frequency, has described sound property to make described second audio data.
Specifically, first obtains unit 202 for analyzing the waveform of described first voice data, extracts described first audio frequency parameter.
And correcting unit 203 is trained the nervous system that described audio frequency exports application program for utilizing the first audio frequency parameter, export described second audio data to make described nervous system.
Carry out timing at correcting unit to nervous system, correcting unit 203, for based on described first audio frequency parameter and described neural current spectral, revises described at least one first parameter neural; Obtain described nervous system export correction after frequency spectrum; After judging described correction, whether the similarity of the frequency spectrum of frequency spectrum and described first voice data equals to be greater than first threshold; When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data equals to be greater than described first threshold, terminate described neural training, and control described nervous system and export described second audio data according to the parameter after correcting; When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data is less than described first threshold, using the current spectral that frequency spectrum after described correction corrects as next round, perform step: based on described first audio frequency parameter and described neural current spectral, revise described at least one first parameter neural.
Optionally, described electronic equipment also comprises:
Statistic unit, for adding up the frequency of each dialogue key element in described first voice data;
Second determining unit, for when the frequency of described dialogue key element exceedes Second Threshold, determines that described dialogue key element is common dialogue key element;
Adding device, for adding to described common dialogue key element in described second audio data.
The various variation pattern of the information processing method in earlier figures 1 embodiment and instantiation are equally applicable to the electronic equipment of the present embodiment, by the aforementioned detailed description to information processing method, those skilled in the art clearly can know the implementation method of electronic equipment in the present embodiment, so succinct in order to instructions, be not described in detail in this.
Above-mentioned one or more technical scheme in the embodiment of the present application, at least has one or more technique effects following:
In the technical scheme of the embodiment of the present application, electronic equipment determines based on first operation of user the first voice data that user selects from multiple voice data, and then from the first voice data, obtain the first audio frequency parameter of the sound property for characterizing the first voice data.Utilize the first audio frequency parameter to correct the second audio data that audio frequency written-out program exports, make second audio data have the sound property of the first voice data.So, solve the technical matters of the sound property that user in prior art can not select voice output application program to export, achieve the technique effect needing that the sound property of voice output application program is corrected according to user.For user, the first voice data of the sound property selecting oneself to need from multiple voice data, such as user likes the sound of host A, then select to speak the voice data of people for host A as the first voice data.Thus electronic equipment is corrected the second audio data that audio frequency exports application program output.User just can hear that audio frequency exports application program host A voice output second audio data.
Those skilled in the art should understand, embodiments of the invention can be provided as method, system or computer program.Therefore, the present invention can adopt the form of complete hardware embodiment, completely software implementation or the embodiment in conjunction with software and hardware aspect.And the present invention can adopt in one or more form wherein including the upper computer program implemented of computer-usable storage medium (including but not limited to magnetic disk memory, CD-ROM, optical memory etc.) of computer usable program code.
The present invention describes with reference to according to the process flow diagram of the method for the embodiment of the present invention, equipment (system) and computer program and/or block scheme.Should understand can by the combination of the flow process in each flow process in computer program instructions realization flow figure and/or block scheme and/or square frame and process flow diagram and/or block scheme and/or square frame.These computer program instructions can being provided to the processor of multi-purpose computer, special purpose computer, Embedded Processor or other programmable data processing device to produce a machine, making the instruction performed by the processor of computing machine or other programmable data processing device produce device for realizing the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.
These computer program instructions also can be stored in can in the computer-readable memory that works in a specific way of vectoring computer or other programmable data processing device, the instruction making to be stored in this computer-readable memory produces the manufacture comprising command device, and this command device realizes the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.
These computer program instructions also can be loaded in computing machine or other programmable data processing device, make on computing machine or other programmable devices, to perform sequence of operations step to produce computer implemented process, thus the instruction performed on computing machine or other programmable devices is provided for the step realizing the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.
Specifically, the computer program instructions that a kind of information processing method in the embodiment of the present application is corresponding can be stored in CD, hard disk, on the storage mediums such as USB flash disk, when the computer program instructions corresponding with the first information processing method in storage medium is read by an electronic equipment or be performed, comprise the steps:
Based on first operation of user, from multiple voice data, determine the first voice data;
From described first voice data, obtain the first audio frequency parameter, wherein, described first audio frequency parameter is for characterizing the sound property of described first voice data;
Utilize described first audio frequency parameter to correct the second audio data being exported application program output by described audio frequency, to make described second audio data, there is described sound property.
Optionally, what store in described storage medium obtains the first audio frequency parameter with step from described first voice data, and corresponding computer instruction, being specifically performed in process, specifically comprises the steps:
The waveform of described first voice data is analyzed, extracts described first audio frequency parameter.
Optionally, what store in described storage medium utilizes described first audio frequency parameter to correct with step to export by described audio frequency the second audio data that application program exports, and corresponding computer instruction, being specifically performed in process, specifically comprises the steps:
Utilize the first audio frequency parameter to train the nervous system that described audio frequency exports application program, export described second audio data to make described nervous system.
Optionally, what store in described storage medium utilizes the first audio frequency parameter to train the nervous system that described audio frequency exports application program with step, described second audio data is exported to make described nervous system, corresponding computer instruction, being specifically performed in process, specifically comprises the steps:
Based on described first audio frequency parameter and described neural current spectral, revise described at least one first parameter neural;
Obtain described nervous system export correction after frequency spectrum;
After judging described correction, whether the similarity of the frequency spectrum of frequency spectrum and described first voice data equals to be greater than first threshold;
When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data equals to be greater than described first threshold, terminate described neural training, and control described nervous system and export described second audio data according to the parameter after correcting;
When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data is less than described first threshold, using the current spectral that frequency spectrum after described correction corrects as next round, perform step: based on described first audio frequency parameter and described neural current spectral, revise described at least one first parameter neural.
Optionally, other computer instruction is also stored in described storage medium, these computer instructions with step: utilize described first audio frequency parameter correct the second audio data exported by described output application program after be performed, comprise the steps: when being performed
Add up the frequency of each dialogue key element in described first voice data;
When the frequency of described dialogue key element exceedes Second Threshold, determine that described dialogue key element is common dialogue key element;
Described common dialogue key element is added in described second audio data.
Obviously, those skilled in the art can carry out various change and modification to the present invention and not depart from the spirit and scope of the present invention.Like this, if these amendments of the present invention and modification belong within the scope of the claims in the present invention and equivalent technologies thereof, then the present invention is also intended to comprise these change and modification.
Claims (10)
1. an information processing method, is applied to electronic equipment, and described electronic equipment can run an audio frequency and export application program, comprising:
Based on first operation of user, from multiple voice data, determine the first voice data;
From described first voice data, obtain the first audio frequency parameter, wherein, described first audio frequency parameter is for characterizing the sound property of described first voice data;
Utilize described first audio frequency parameter to correct the second audio data being exported application program output by described audio frequency, to make described second audio data, there is described sound property.
2. the method for claim 1, is characterized in that, obtains the first audio frequency parameter, comprising from described first voice data:
The waveform of described first voice data is analyzed, extracts described first audio frequency parameter.
3. method as claimed in claim 2, is characterized in that, utilizes described first audio frequency parameter to correct the second audio data being exported application program output by described audio frequency, comprising:
Utilize the first audio frequency parameter to train the nervous system that described audio frequency exports application program, export described second audio data to make described nervous system.
4. method as claimed in claim 3, is characterized in that, utilizes the first audio frequency parameter to train the nervous system that described audio frequency exports application program, comprising:
Based on described first audio frequency parameter and described neural current spectral, revise described at least one first parameter neural;
Obtain described nervous system export correction after frequency spectrum;
After judging described correction, whether the similarity of the frequency spectrum of frequency spectrum and described first voice data equals to be greater than first threshold;
When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data equals to be greater than described first threshold, terminate described neural training, and control described nervous system and export described second audio data according to the parameter after correcting;
When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data is less than described first threshold, using the current spectral that frequency spectrum after described correction corrects as next round, perform step: based on described first audio frequency parameter and described neural current spectral, revise described at least one first parameter neural.
5. method as claimed in claim 3, is characterized in that, after the second audio data utilizing described first audio frequency parameter correction to be exported by described output application program, described method also comprises:
Add up the frequency of each dialogue key element in described first voice data;
When the frequency of described dialogue key element exceedes Second Threshold, determine that described dialogue key element is common dialogue key element;
Described common dialogue key element is added in described second audio data.
6. an electronic equipment, described electronic equipment can run an audio frequency and export application program, and described electronic equipment comprises:
First determining unit, for the first operation based on user, determines the first voice data from multiple voice data;
First obtains unit, and for obtaining the first audio frequency parameter from described first voice data, wherein, described first audio frequency parameter is for characterizing the sound property of described first voice data;
Correcting unit, for utilizing described first audio frequency parameter to correct the second audio data being exported application program output by described audio frequency, has described sound property to make described second audio data.
7. electronic equipment as claimed in claim 6, is characterized in that, described first obtains unit is used for analyzing the waveform of described first voice data, extracts described first audio frequency parameter.
8. electronic equipment as claimed in claim 7, is characterized in that, described correcting unit is trained the nervous system that described audio frequency exports application program for utilizing the first audio frequency parameter, exports described second audio data to make described nervous system.
9. electronic equipment as claimed in claim 8, is characterized in that, described correcting unit is used for based on described first audio frequency parameter and described neural current spectral, revises described at least one first parameter neural; Obtain described nervous system export correction after frequency spectrum; After judging described correction, whether the similarity of the frequency spectrum of frequency spectrum and described first voice data equals to be greater than first threshold; When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data equals to be greater than described first threshold, terminate described neural training, and control described nervous system and export described second audio data according to the parameter after correcting; When the similarity of the frequency spectrum of frequency spectrum after described correction and described first voice data is less than described first threshold, using the current spectral that frequency spectrum after described correction corrects as next round, perform step: based on described first audio frequency parameter and described neural current spectral, revise described at least one first parameter neural.
10. electronic equipment as claimed in claim 9, it is characterized in that, described electronic equipment also comprises:
Statistic unit, for adding up the frequency of each dialogue key element in described first voice data;
Second determining unit, for when the frequency of described dialogue key element exceedes Second Threshold, determines that described dialogue key element is common dialogue key element;
Adding device, for adding to described common dialogue key element in described second audio data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510583838.2A CN105206280A (en) | 2015-09-14 | 2015-09-14 | Information processing method and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510583838.2A CN105206280A (en) | 2015-09-14 | 2015-09-14 | Information processing method and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105206280A true CN105206280A (en) | 2015-12-30 |
Family
ID=54953909
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510583838.2A Pending CN105206280A (en) | 2015-09-14 | 2015-09-14 | Information processing method and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105206280A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106057208A (en) * | 2016-06-14 | 2016-10-26 | 科大讯飞股份有限公司 | Audio correction method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120253794A1 (en) * | 2011-03-29 | 2012-10-04 | Kabushiki Kaisha Toshiba | Voice conversion method and system |
| CN104424952A (en) * | 2013-08-20 | 2015-03-18 | 索尼公司 | Voice processing apparatus, voice processing method, and program |
| CN104538011A (en) * | 2014-10-30 | 2015-04-22 | 华为技术有限公司 | Tone adjusting method and device and terminal device |
| CN104575487A (en) * | 2014-12-11 | 2015-04-29 | 百度在线网络技术（北京）有限公司 | Voice signal processing method and device |
| CN104636111A (en) * | 2015-02-15 | 2015-05-20 | 联想(北京)有限公司 | Information processing method and electronic device |
-
2015
- 2015-09-14 CN CN201510583838.2A patent/CN105206280A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120253794A1 (en) * | 2011-03-29 | 2012-10-04 | Kabushiki Kaisha Toshiba | Voice conversion method and system |
| CN104424952A (en) * | 2013-08-20 | 2015-03-18 | 索尼公司 | Voice processing apparatus, voice processing method, and program |
| CN104538011A (en) * | 2014-10-30 | 2015-04-22 | 华为技术有限公司 | Tone adjusting method and device and terminal device |
| CN104575487A (en) * | 2014-12-11 | 2015-04-29 | 百度在线网络技术（北京）有限公司 | Voice signal processing method and device |
| CN104636111A (en) * | 2015-02-15 | 2015-05-20 | 联想(北京)有限公司 | Information processing method and electronic device |
Non-Patent Citations (2)
| Title |
|---|
| 张军英: "《说话人识别的现代方法与技术》", 31 October 1994, 西北大学出版社 * |
| 张洪润: "《传感器技术大全（下期）》", 31 October 2007, 北京航空航天大学出版社 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106057208A (en) * | 2016-06-14 | 2016-10-26 | 科大讯飞股份有限公司 | Audio correction method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109801644B (en) | Separation method, separation device, electronic equipment and readable medium for mixed sound signal | |
| CN107644642B (en) | Semantic recognition method and device, storage medium and electronic equipment | |
| CN109326289A (en) | Exempt to wake up voice interactive method, device, equipment and storage medium | |
| CN102568478B (en) | Video play control method and system based on voice recognition | |
| EP3779972A1 (en) | Voice wake-up method and apparatus | |
| CN107977183A (en) | voice interactive method, device and equipment | |
| CN105976812A (en) | Voice identification method and equipment thereof | |
| WO2017166650A1 (en) | Voice recognition method and device | |
| US10860289B2 (en) | Flexible voice-based information retrieval system for virtual assistant | |
| JP6783339B2 (en) | Methods and devices for processing audio | |
| CN108877790A (en) | Speaker control method, device, readable storage medium storing program for executing and mobile terminal | |
| CN104123938A (en) | Voice control system, electronic device and voice control method | |
| CN107293300A (en) | Audio recognition method and device, computer installation and readable storage medium storing program for executing | |
| CN109979450B (en) | Information processing method and device and electronic equipment | |
| CN104866275B (en) | Method and device for acquiring image information | |
| CN112309365A (en) | Training method and device of speech synthesis model, storage medium and electronic equipment | |
| CN105469789A (en) | Voice information processing method and voice information processing terminal | |
| CN107886963B (en) | A kind of method, apparatus and electronic equipment of speech processes | |
| CN110277092A (en) | A kind of voice broadcast method, device, electronic equipment and readable storage medium storing program for executing | |
| CN110444190A (en) | Method of speech processing, device, terminal device and storage medium | |
| CN111145730A (en) | Method and system for optimizing speech recognition model | |
| CN110688518A (en) | Rhythm point determining method, device, equipment and storage medium | |
| CN108874363A (en) | Object control method, apparatus, equipment and storage medium for AR scene | |
| CN110503968A (en) | A kind of audio-frequency processing method, device, equipment and readable storage medium storing program for executing | |
| CN111145732A (en) | Processing method and system after multi-task voice recognition |
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: 20151230 |
|
| RJ01 | Rejection of invention patent application after publication |