CN110111816A - Method, the method for audio processing, electronic equipment and the server-side of recording audio - Google Patents
Method, the method for audio processing, electronic equipment and the server-side of recording audio Download PDFInfo
- Publication number
- CN110111816A CN110111816A CN201910147012.XA CN201910147012A CN110111816A CN 110111816 A CN110111816 A CN 110111816A CN 201910147012 A CN201910147012 A CN 201910147012A CN 110111816 A CN110111816 A CN 110111816A
- Authority
- CN
- China
- Prior art keywords
- audio
- recording
- fragment
- target
- audio fragment
- 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
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/18—Speech classification or search using natural language modelling
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
- G11B20/00—Signal processing not specific to the method of recording or reproducing; Circuits therefor
- G11B20/10—Digital recording or reproducing
- G11B20/10527—Audio or video recording; Data buffering arrangements
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
- G11B20/00—Signal processing not specific to the method of recording or reproducing; Circuits therefor
- G11B20/10—Digital recording or reproducing
- G11B20/10527—Audio or video recording; Data buffering arrangements
- G11B2020/10537—Audio or video recording
- G11B2020/10546—Audio or video recording specifically adapted for audio data
Abstract
The present embodiments relate to fields of communication technology, disclose method, the method for audio processing, electronic equipment and the server-side of a kind of recording audio.The method of recording audio in the present invention is applied to recording device, comprising: obtains the semantics recognition of the audio fragment currently acquired as a result, and determining the affiliated theme of audio fragment according to semantics recognition result；Judge whether the affiliated theme of audio fragment and target topic are identical, and according to judging result, control the recording process of audio, wherein target topic is the generic to recorded audio content.Present embodiment can automatically control the recording to audio signal, avoid the typing of invalid audio fragment, improve the quality of recording.
Description
Technical field
The present embodiments relate to field of communication technology, in particular to the side of a kind of method of recording audio, audio processing
Method, electronic equipment and server-side.
Background technique
With the continuous development of science and technology, current electronic equipment usually all has sound-recording function, for example, smart phone has
Sound-recording function, MP3 have sound-recording function etc..User generallys use the equipment with sound-recording function in conference content or interview
Appearance is recorded.
When user is recorded using the sound pick-up outfit, user can click " start button " on the sound pick-up outfit, with
Start the recording that sound pick-up outfit carries out audio, and at the end of the audio recording, then user needs to click conclusion button (or again
Click start button) to terminate audio recording.
At least there are the following problems in the prior art for inventor's discovery: at present in recording audio, user being needed to open manually
Audio recording is turned off manually in dynamic audio recording.But in actual use, since usage scenario is complicated and changeable, user is often not
Recording (or even many times user often forgets that pause is recorded), the invalid audio of typing, to reduce can be suspended in time
Recording quality, while the invalid audio of typing can also occupy the memory space of sound pick-up outfit.For example, user B starting recording is set
It is standby, the interview content with A is recorded, A has been connected to a phone suddenly during recording, and A is directed to the conversation content of phone simultaneously
It is not the audio content that this interview needs to record, needs user B to suspend recording manually at this time, if user B does not suspend in time
Record or forget pause record, then can the invalid audio of typing reduced since message is uncorrelated to this recording substance
Recording quality increases the time (when arranging interview data, needing to delete invalid audio) that subsequent user arranges recording substance, together
Shi Wuxiao audio also takes up the memory space of sound pick-up outfit.
Summary of the invention
A kind of method for being designed to provide recording of embodiment of the present invention allows to automatically control to audio signal
Recording, avoid the typing of invalid audio fragment, improve the quality of recording.
In order to solve the above technical problems, embodiments of the present invention provide a kind of method of recording audio, it is applied to record
Mixer, comprising: obtain the semantics recognition of the audio fragment currently acquired as a result, and determining audio piece according to semantics recognition result
Theme belonging to section；Judge whether the affiliated theme of audio fragment and target topic are identical, and according to judging result, controls the record of audio
Process processed, wherein target topic is the generic to recorded audio content.
Embodiments of the present invention additionally provide a kind of method of audio processing, are applied to server-side, comprising: receive recording
The audio data that device is sent, wherein audio data includes the target audio that recording device is recorded and the control for recording target audio
Information processed, information at the time of control information includes pause recording audio；According to the control information, target audio is carried out at editing
Reason；Wherein, recording device records the process of target audio are as follows: is determined according to the semantics recognition result of the audio fragment currently acquired
The affiliated theme of audio fragment judges whether the affiliated theme of audio fragment and target topic are identical, according to judging result, controls sound
The recording process of frequency.
Embodiments of the present invention additionally provide a kind of electronic equipment, comprising: at least one processor；And at least
The memory of one processor communication connection；Wherein, memory is stored with the instruction that can be executed by least one processor, instruction
It is executed by least one processor, so that the method that at least one processor is able to carry out above-mentioned recording audio.
Embodiments of the present invention additionally provide a kind of server-side, comprising: at least one processor；And at least one
The memory of a processor communication connection；Wherein, memory is stored with the instruction that can be executed by least one processor, instructs quilt
At least one processor executes, so that the method that at least one processor is able to carry out above-mentioned audio processing.
Embodiment of the present invention in terms of existing technologies, by by the affiliated theme of the audio fragment currently acquired and mesh
Mark theme is compared, and the recording process of audio is controlled according to judging result, and audio recording process includes to audio recording
Pause and restart audio recording；Due to during the entire process of recording audio, without suspending the record of audio manually
System, the problem of typing audio fragment unrelated with target topic due to manually forgetting pause can be avoided the occurrence of, to improve record
The quality of sound.And semantics recognition is used to the audio fragment of acquisition, it can determine the content that the audio fragment is recorded, and then can be with
The affiliated theme of the audio fragment currently acquired is quickly and accurately determined according to the content of recording；Simultaneously as in entire audio
In recording process, be by judge acquisition the affiliated theme of audio fragment and target topic whether unanimously control audio record
System can also be avoided the occurrence of because manually forget to cause to miss due to cancelling pause recording relevant to target topic audio the problem of,
The step of further improving the quality of recording audio, reducing processing of the later period to the audio of recording, improves the audio to recording
Processing speed.
In addition, the method for recording audio further include: if receiving the instruction that instruction terminates the recording of audio, terminate to sound
The recording of frequency, obtains target audio；Audio data is uploaded to server-side, audio data includes target audio and recording target
The control information of audio, wherein information at the time of control information includes pause recording audio.By by target audio and recording
The control information of target audio is uploaded to server-side, can be handled by controlling information target audio by server-side, e.g.,
Editing processing, can simplify the step of follow-up service end is to audio processing, accelerate the speed to audio processing, to pass through service
The audio quality of target audio is improved again in end.
In addition, controlling the recording process of audio according to judging result, specifically including: if it is determined that the affiliated theme of audio fragment
It is not identical as target topic, then suspend the recording of audio；In the recording process of pause audio, if detecting and target topic phase
Same audio fragment, then recording of the restarting to audio.It include to the temporary of audio recording during controlling audio recording
Stop and restart, suspends the recording of audio, it can be to avoid typing and the incoherent audio fragment of target topic；And in pause audio
In recording process, if detecting, the affiliated theme of the audio fragment of acquisition is identical as target topic, restarts the recording of audio,
The problem of missing recording identical with target topic audio fragment can be avoided the occurrence of.
In addition, obtaining the semantics recognition of the audio fragment currently acquired as a result, specifically including: the audio piece that will currently acquire
Section is uploaded to server-side, and receives the semantics recognition result of the audio fragment fed back by server-side；Alternatively, to the sound currently acquired
Frequency segment carries out semantics recognition, obtains the semantics recognition result of audio fragment.It provides two kinds and obtains the audio fragment currently acquired
Semantics recognition result mode, convenient for flexibly obtaining the semantics recognition result of the audio fragment currently acquired.
In addition, before the semantics recognition result for obtaining the audio fragment currently acquired, the method for recording audio further include:
The semantics recognition of the first section audio fragment acquired for the first time is obtained as a result, and according to the semantics recognition of first section audio fragment as a result, really
Determine the affiliated theme of first section audio fragment；And using the affiliated theme of first section audio fragment as target topic.Due to first section audio fragment
In generally comprise subject content to recording audio, can be with thus using the affiliated theme of first section audio fragment as target topic
Quickly and accurately determine target topic, and implementation is simple.
In addition, in the recording process of pause audio, the method for recording audio further include: in the recording process of pause audio
In, if detecting audio fragment identical with target topic, save audio fragment identical with target topic.In pause audio
In recording process, save identical with target topic audio fragment, avoid restart record when, occur leakage preservation currently with
The problem of target topic identical audio fragment, to improve the quality of audio recording.
Detailed description of the invention
One or more embodiments are illustrated by the picture in corresponding attached drawing, these exemplary theorys
The bright restriction not constituted to embodiment, the element in attached drawing with same reference numbers label are expressed as similar element, remove
Non- to have special statement, composition does not limit the figure in attached drawing.
Fig. 1 is a kind of method idiographic flow schematic diagram for recording audio that first embodiment provides according to the present invention；
Fig. 2 is a kind of method idiographic flow schematic diagram for recording audio that second embodiment provides according to the present invention；
Fig. 3 is a kind of method idiographic flow schematic diagram for audio processing that third embodiment provides according to the present invention；
Fig. 4 is a kind of method idiographic flow schematic diagram for audio processing that the 4th embodiment provides according to the present invention；
Fig. 5 is the concrete structure schematic diagram for a kind of electronic equipment that the 5th embodiment provides according to the present invention；
Fig. 6 is a kind of concrete structure schematic diagram for server-side that sixth embodiment provides according to the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Each embodiment be explained in detail.However, it will be understood by those skilled in the art that in each embodiment party of the present invention
In formula, in order to make the reader understand this application better, many technical details are proposed.But even if without these technical details
And various changes and modifications based on the following respective embodiments, the application technical solution claimed also may be implemented.
The division of each embodiment is for convenience, should not to constitute to specific implementation of the invention any below
It limits, each embodiment can be combined with each other mutual reference under the premise of reconcilable.
The first embodiment of the present invention is related to a kind of methods of recording audio.The method of the recording audio is applied to recording
Device, the recording device can be the electronic equipment with sound-recording function, as: smart phone, recording pen, with sound-recording function
MP3 etc..The detailed process of the method for the recording audio is as shown in Figure 1.
Step 101: obtaining the semantics recognition of the audio fragment currently acquired as a result, and determining sound according to semantics recognition result
The affiliated theme of frequency segment.
Specifically, start the recording device, which acquires the audio letter of the recording device ambient enviroment in real time
Number, such as: the talk audio signal of user.Know wherein it is possible to obtain the semantic of the audio fragment currently acquired according to predetermined period
Not as a result, for example, can just obtain the semantics recognition result of an audio fragment every 30 seconds.It can not also be according to predetermined period
The semantics recognition of the audio fragment currently acquired is obtained as a result, for example, obtaining primary if detecting in 2 seconds without audio input
The semantics recognition of the audio fragment currently acquired as a result, at the time of the audio fragment currently acquired is last obtains to it is current when
Audio fragment in quarter.
The corresponding keyword of each theme can be preset, for example, theme is that keyword corresponding to " film " can be with
Including " box office ", " showing ", " attendance ", " screening ", " play ", " director " and " protagonist " etc., theme " game " is corresponding
Keyword may include: " E3 ", " Sony ", " big method ", " Nintendo ", " 3A " and " Steam " etc.；Theme is " finance " institute
Corresponding keyword may include: " currency ", " trade war ", " fund ", " security ", " price of gold " and " stock market " etc..Certainly, often
The quantity of keyword corresponding to a theme is with no restrictions.
The keyword in semantics recognition result is got, keyword corresponding to each theme in the audio fragment is counted
Quantity chooses the theme comprising most keywords as the affiliated theme of the audio fragment.For example, the audio fragment currently acquired
Semantics recognition result are as follows: " the big method of this E3 Sony is severe！There are so more exclusive 3A your writings！", by the semantics recognition result
In word be compared with keyword corresponding to each theme, can determine occur 4 and theme in the audio fragment
" game " relevant keyword (i.e. " E3 ", " Sony ", " big method " and " 3A "), that is, can determine that the affiliated theme of the audio fragment is
" game ".
It is noted that needing to obtain target before the semantics recognition result for obtaining the audio fragment currently acquired
Theme, there are many modes for obtaining target topic, for example, suggestion voice can be exported, prompting user's input, (input mode can be with
It is voice input, keyboard input text etc.) target topic；It can also be obtained automatically by way of the automatic semantics recognition of recording device
Take target topic.The automatic mode for obtaining target topic is described below:
The semantics recognition of the first section audio fragment acquired for the first time is obtained as a result, and according to the semantics recognition of first section audio fragment
As a result, determining the affiliated theme of first section audio fragment；And using the affiliated theme of first section audio fragment as target topic.
Specifically, after first section audio fragment can be recording device starting, the first segment audio fragment of acquisition, and obtain
The semantics recognition of the first section audio fragment as a result, extract the keyword in the semantics recognition result of the first section audio fragment, and with
Keyword corresponding to preset themes is compared, so that it is determined that going out the affiliated theme of first section audio fragment, for example, first section audio
The semantics recognition result of segment is " me is next allowed to chat middle-east situation ", can extract keyword " middle-east situation ", the pass
The corresponding theme of keyword is " middle-east situation ", that is, can determine that the affiliated theme of first section audio fragment is " middle-east situation ".
In one concrete implementation, there are many modes of the semantics recognition result for the audio fragment that acquisition currently acquires, this
Embodiment is using the two ways being exemplified below.
Mode one: the audio fragment currently acquired is uploaded to server-side, and receives the audio fragment fed back by server-side
Semantics recognition result.
Specifically, the audio fragment currently acquired is uploaded to server-side, audio of the server-side to upload by recording device
Segment carries out semantics recognition, can be using automatic speech recognition method (Automatic Speech Recognition, abbreviation
" ASR "), recording device receives the semantics recognition result of the audio fragment of server-side return.
Mode two: semantics recognition is carried out to the audio fragment currently acquired, obtains the semantics recognition result of audio fragment.
Recording device oneself directly can also carry out semantics recognition to the audio fragment currently acquired, to obtain the audio
The semantics recognition result of segment.
Both the above acquisition modes can be selected according to actual needs, for example, networking in recording device and server-side
In the case where, the semantics recognition of the audio fragment currently acquired can be obtained with pass-through mode one as a result, if recording device is in nothing
In the case where net, employing mode two obtains the semantics recognition result of the audio fragment currently acquired.
Step 102: judging whether the affiliated theme of audio fragment and target topic are identical, and according to judging result, control sound
The recording process of frequency, wherein target topic is the generic to recorded audio content.
In one concrete implementation, according to judging result, controlling audio recording process includes: if it is determined that belonging to audio fragment
Theme is not identical as target topic, then suspends the recording of audio；In the recording process of pause audio, if detecting and target master
Identical audio fragment is inscribed, then recording of the restarting to audio.
Specifically, however, it is determined that the affiliated theme of audio fragment is identical as target topic, then continues the recording of audio.If it is determined that
The affiliated theme of audio fragment and target topic be not identical, then suspends the recording of audio, and carry out to the audio fragment acquired in real time
Detection, whether the affiliated theme of audio fragment that detection acquires in real time is identical as target topic, if detecting the audio currently acquired
The affiliated theme of segment is identical as target topic, then recording of the restarting to audio.
It is noted that step 101 and step 102 can be repeated during the entire process of recording audio, thus
Control the recording to entire audio.
Embodiment of the present invention in terms of existing technologies, by by the affiliated theme of the audio fragment currently acquired and mesh
Mark theme is compared, and the recording process of audio is controlled according to judging result, and audio recording process includes to audio recording
Pause and restart audio recording；Due to during the entire process of recording audio, without suspending the record of audio manually
System, the problem of typing audio fragment unrelated with target topic due to manually forgetting pause can be avoided the occurrence of, to improve record
The quality of sound.And semantics recognition is used to the audio fragment of acquisition, it can determine the content that the audio fragment is recorded, and then can be with
The affiliated theme of the audio fragment currently acquired is quickly and accurately determined according to the content of recording；Simultaneously as in entire audio
In recording process, be by judge acquisition the affiliated theme of audio fragment and target topic whether unanimously control audio record
System can also be avoided the occurrence of because manually forget to cause to miss due to cancelling pause recording relevant to target topic audio the problem of,
The step of further improving the quality of recording audio, reducing processing of the later period to the audio of recording, improves the audio to recording
Processing speed.
Second embodiment of the present invention is related to a kind of method of recording audio.Second embodiment is to the first embodiment party
The further improvement of formula, mainly thes improvement is that: in second embodiment of the invention, if receiving instruction terminates audio
The instruction of recording then terminates recording to audio, obtains target audio, and by the audio data upload service comprising target audio
End.The detailed process of the method for the recording audio is as shown in Figure 2.
Step 201: obtaining the semantics recognition of the audio fragment currently acquired as a result, and determining sound according to semantics recognition result
The affiliated theme of frequency segment.
Step 202: judging whether the affiliated theme of audio fragment and target topic are identical, and according to judging result, control sound
The recording process of frequency, wherein target topic is the generic to recorded audio content.
One in the specific implementation, during suspending audio recording, if detecting audio fragment identical with target topic,
Save audio fragment identical with target topic.
Specifically, during suspending audio recording, recording device enters listening mode, i.e., obtains the sound of acquisition in real time
The semantics recognition of frequency segment is as a result, judge whether the affiliated theme of audio fragment of acquisition is identical as target topic；For the ease of obtaining
Take the semantics recognition to the audio fragment currently acquired as a result, can be in the affiliated theme of audio fragment for judging currently to acquire every time
With target topic it is whether consistent before, cache the audio fragment, however, it is determined that the affiliated theme of the audio fragment is identical as target topic,
The recording to audio is restarted, meanwhile, save the audio fragment, it is ensured that recording audio identical with target topic will not be missed
Segment；If it is determined that the affiliated theme of the audio fragment and target topic be not identical, then the audio fragment is not saved.
In the recording process of pause audio, audio fragment identical with target topic is saved, avoids recording in restarting
When processed, there is the problem of leakage saves currently audio fragment identical with target topic, to improve the quality of audio recording.
Step 203: if receiving the instruction that instruction terminates the recording of audio, terminating the recording to audio, obtain target
Audio.
Specifically, user can send the instruction for terminating audio recording to the recording device, which can be by pre-
If operation input, for example, inputting instruction by the end key on recording device, or (such as: terminating by speech-input instructions
It records).Recording device upon receipt of the instructions, terminates the recording to audio, and what is obtained is target audio.
Step 204: audio data being uploaded to server-side, audio data includes target audio and recording target audio
Control information, wherein information at the time of control information includes pause recording audio.
Specifically, information at the time of control information includes pause recording audio, for example, control information includes pause audio
Recording the t1 moment, and restarting is to the t3 moment of the recording of audio, wherein the t1 moment is earlier than the t3 moment.By target
Audio and control information as audio data are uploaded to server-side, are cut according to the control information to target audio by server-side
Processing is collected, for example, server-side can obtain and suspend audio fragment corresponding at the time of recording audio according to the control information,
Corresponding audio fragment, determines the identity characteristic information of invalidated object at the time of according to pause recording audio；Based on invalid right
The identity characteristic information of elephant determines the corresponding audio fragment of invalidated object；Using the corresponding audio fragment of invalidated object as invalid sound
Frequency segment, and delete invalid audio fragment.
It should be noted that step 201, step 202 in present embodiment respectively with the step in first embodiment
101 and step 102 it is roughly the same, will no longer repeat herein.
The method of the recording audio provided in present embodiment, by by target audio and record target audio control
Information is uploaded to server-side, can carry out editing processing to target audio by control information by server-side, can simplify subsequent
The step of server-side is to audio processing accelerates the speed to audio processing, to improve target audio again by server-side
Audio quality.
The step of various methods divide above, be intended merely to describe it is clear, when realization can be merged into a step or
Certain steps are split, multiple steps are decomposed into, as long as including identical logical relation, all in the protection scope of this patent
It is interior；To adding inessential modification in algorithm or in process or introducing inessential design, but its algorithm is not changed
Core design with process is all in the protection scope of the patent.
Third embodiment of the invention is related to a kind of method of audio processing, and the method for the audio processing is applied to service
End, server-side and recording device communicate to connect, and server-side can be in communication with each other with recording device, the server-side can be cloud,
Server etc..The detailed process of the method for the audio processing is as shown in Figure 3.
Step 301: receiving the audio data that recording device is sent, wherein audio data includes the mesh that recording device is recorded
Mark with phonetic symbols frequency and the control information for recording target audio, information at the time of control information includes pause recording audio.
In one concrete implementation, recording device records the process of target audio are as follows: according to the audio fragment currently acquired
Semantics recognition result determine the affiliated theme of audio fragment, judge whether the affiliated theme of audio fragment and target topic identical,
According to judging result, the recording process of audio is controlled.
Recording device controls the recording process of audio according to judging result, after recording device terminates to audio recording, i.e.,
Target audio can be obtained.Recording device is using obtained target audio and records the control information of the target audio as audio data
Upload service end, server-side receive the audio data.
Step 302: according to the control information, editing processing being carried out to target audio.
In one concrete implementation, according to information at the time of pause recording audio, at the time of acquisition with pause recording audio
Corresponding audio fragment；Corresponding audio fragment at the time of according to pause recording audio, determines that the identity of invalidated object is special
Reference breath；The corresponding audio fragment of the invalidated object is determined based on the identity characteristic information of the invalidated object；By the invalidated object
Corresponding audio fragment deletes invalid audio fragment as invalid audio fragment.
Specifically, for the ending with the different audio fragment of target topic at the time of suspending recording audio, one temporarily
Stop having a corresponding audio fragment at the time of recording audio.Recording device is right always before pause is to the operation of audio recording
Audio fragment is saved, thus, recording device can save audio fragment corresponding to the pause moment.
According to the pause moment can obtain with audio fragment corresponding at the time of suspending recording audio, suspend recording audio
The moment affiliated theme of corresponding audio fragment is not identical as target topic, obtains corresponding audio at the time of the pause recording audio
The identity characteristic information of invalidated object in segment, identity characteristic information can be tone color, tone etc., and then according to the invalidated object
Identity characteristic information.The identity characteristic information of the invalidated object is compared in entire target audio, determines have
The audio fragment of the identity characteristic information of the invalidated object, using the corresponding audio fragment of invalidated object as invalid audio fragment,
The invalid audio fragment determined is deleted from the target audio.
It is noted that after server-side carries out editing processing to target audio, it can be by editing treated target sound
Frequency feeds back to recording device.
The method of the audio processing provided in present embodiment is determined to need by the control information in audio data
The audio fragment of editing, to realize that the automatic editing to target audio simplifies the step of audio processing without human intervention
Suddenly, the speed to audio processing is improved.In addition, according to the control information, the audio fragment of invalidated object is determined, to the target sound
Invalid audio fragment is deleted in frequency, further improves the audio quality of target audio.
Four embodiment of the invention is related to a kind of method of audio processing.4th embodiment is to third embodiment
Further improvement, mainly the improvement is that: according to the control information, after carrying out editing processing to target audio, the sound
Frequency processing method can also to editing, treated that target audio is handled according to target topic, specific process such as Fig. 4 institute
Show.
Step 401: receiving the audio data that recording device is sent, wherein audio data includes the mesh that recording device is recorded
Mark with phonetic symbols frequency and the control information for recording target audio, information at the time of control information includes pause recording audio.
Step 402: according to the control information, editing processing being carried out to target audio.
Step 403: obtaining the affiliated theme of first section audio fragment, and as target topic.
Specifically, semantics recognition directly is carried out to the first section audio of target audio, obtains the semantic of the first section audio and knows
Not as a result, determining the affiliated theme of first section audio fragment according to the keyword in the semantics recognition result, and as target master
Topic.
Step 404: the target audio in addition to first section audio fragment is split as N number of audio fragment, N is the integer greater than 1,
And each audio fragment is handled.
Specifically, segment processing is carried out to target audio according to preset frequency, i.e., will removes first section in the target audio
Audio except audio fragment splits into several audio fragments according to preset frequency, and preset frequency can be according to practical need
It is configured.
In one concrete implementation, to the treatment process of each audio fragment progress are as follows: carry out semantic knowledge to audio fragment
Not, the semantics recognition result of audio fragment is obtained；The affiliated theme of audio fragment is determined according to the semantics recognition result of audio fragment；
The affiliated theme of audio fragment is compared with target topic, however, it is determined that the affiliated theme of audio fragment and target topic be not identical,
Then delete audio fragment.
Semantics recognition is carried out to each audio fragment, obtains recognition result, and obtain the keyword in semantics recognition result；
According to the corresponding relationship between theme and keyword, every affiliated theme of section audio segment is determined, by every section audio segment institute owner
Topic is compared with target topic respectively, will delete from target audio with the different audio fragment of target topic.
The method of the audio processing provided in present embodiment carries out at editing target audio according to the control information
Reason and then it is secondary according to target topic, to editing, treated that target audio is handled, delete different with target topic
Audio fragment further increases the audio quality of target audio.
Fifth embodiment of the invention is related to a kind of electronic equipment, and the specific structure of the electronic equipment 50 is as shown in figure 5, packet
It includes: at least one processor 501；And the memory 502 with the communication connection of at least one processor 501；Wherein, memory
502 are stored with the instruction that can be executed by least one processor 501, and instruction is executed by least one processor 501, so that at least
The method that one processor 501 is able to carry out recording audio in first embodiment or second embodiment.
Present embodiment is entity device embodiment corresponding with first embodiment or second embodiment, this implementation
Mode can work in coordination implementation with first embodiment or second embodiment.It is mentioned in first embodiment or second embodiment
The relevant technical details arrived are still effective in the present embodiment, and in order to reduce repetition, which is not described herein again.
Sixth embodiment of the invention is related to a kind of server-side, and the specific structure of the server-side 60 is as shown in Figure 6, comprising:
At least one processor 601；And the memory 602 with the communication connection of at least one processor 601；Wherein, memory 602
It is stored with the instruction that can be executed by least one processor 601, instruction is executed by least one processor 601, so that at least one
The method that a processor 601 is able to carry out third embodiment or the processing of the 4th embodiment sound intermediate frequency.
Present embodiment is entity device embodiment corresponding with third embodiment or the 4th embodiment, this implementation
Mode can work in coordination implementation with third embodiment or the 4th embodiment.It is mentioned in third embodiment or the 4th embodiment
The relevant technical details arrived are still effective in the present embodiment, and in order to reduce repetition, which is not described herein again.
It is noted that depositing in the electronic equipment in the 5th embodiment and the server-side in sixth embodiment
Reservoir is all made of bus mode with processor and connects, and bus may include the bus and bridge of any number of interconnection, and bus is by one
The various circuits of a or multiple processors and memory link together.Bus can also will such as peripheral equipment, voltage-stablizer and
Various other circuits of management circuit or the like link together, and these are all it is known in the art, therefore, herein not
It is described further again.Bus interface provides interface between bus and transceiver.Transceiver can be an element,
It is also possible to multiple element, such as multiple receivers and transmitter, provides for logical with various other devices over a transmission medium
The unit of letter.The data handled through processor are transmitted on the radio medium by antenna, and further, antenna also receives data
And transfer data to processor.
Processor is responsible for managing bus and common processing, can also provide various functions, including periodically, peripheral interface,
Voltage adjusting, power management and other control functions.And memory can be used for storage processor and execute operation when institute
The data used.
It will be appreciated by those skilled in the art that implementing the method for the above embodiments is that can pass through
Program is completed to instruct relevant hardware, which is stored in a storage medium, including some instructions are used so that one
A equipment (can be single-chip microcontroller, chip etc.) or processor (processor) execute each embodiment the method for the application
All or part of the steps.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.
It will be understood by those skilled in the art that the respective embodiments described above are to realize specific embodiments of the present invention,
And in practical applications, can to it, various changes can be made in the form and details, without departing from the spirit and scope of the present invention.
Claims (10)
1. a kind of method of recording audio, which is characterized in that be applied to recording device, comprising:
The semantics recognition of the audio fragment currently acquired is obtained as a result, and determining the audio piece according to the semantics recognition result
Theme belonging to section；
Judge whether the affiliated theme of the audio fragment and target topic are identical, and according to judging result, controls the recording of audio
Process, wherein the target topic is the generic to recorded audio content.
2. the method for recording audio according to claim 1, which is characterized in that the method for the recording audio further include:
If receiving instruction terminates the instruction of recording of audio, the recording to audio is terminated, target audio is obtained；
Audio data is uploaded to server-side, the audio data includes the target audio and the recording target audio
Control information, wherein information at the time of the control information includes pause recording audio.
3. the method for recording audio according to claim 1 or 2, which is characterized in that according to the judging result, control sound
The recording process of frequency, specifically includes:
If it is determined that the affiliated theme of audio fragment and target topic be not identical, then suspend the recording of audio；
In the recording process of pause audio, if detecting audio fragment identical with the target topic, restarting pair
The recording of audio.
4. the method for recording audio according to claim 1, which is characterized in that in the audio piece for obtaining and currently acquiring
Before the semantics recognition result of section, the method for the recording audio further include:
The semantics recognition of the first section audio fragment acquired for the first time is obtained as a result, and according to the semantics recognition of the first section audio fragment
As a result, determining the affiliated theme of first section audio fragment；
And using the affiliated theme of first section audio fragment as the target topic.
5. the method for recording audio according to claim 3, which is characterized in that the method for the recording audio further include:
In the recording process of pause audio, if detecting audio fragment identical with the target topic, save and the mesh
Mark the identical audio fragment of theme.
6. a kind of method of audio processing, which is characterized in that be applied to server-side, comprising:
Receive recording device send audio data, wherein the audio data include recording device record target audio and
Record the control information of the target audio, information at the time of the control information includes pause recording audio；
According to the control information, editing processing is carried out to the target audio；
Wherein, the recording device records the process of target audio are as follows: according to the semantics recognition knot of the audio fragment currently acquired
Fruit determines the affiliated theme of the audio fragment, judges whether the affiliated theme of the audio fragment and target topic are identical, according to
Judging result controls the recording process of audio.
7. the method for audio processing according to claim 6, which is characterized in that according to the control information, to the mesh
Mark with phonetic symbols frequency carries out editing processing, specifically includes:
According to information at the time of pause recording audio, obtain and audio fragment corresponding at the time of pause recording audio；
Corresponding audio fragment, determines the identity characteristic information of invalidated object at the time of according to pause recording audio；
The corresponding audio fragment of the invalidated object is determined based on the identity characteristic information of the invalidated object；
Using the corresponding audio fragment of the invalidated object as invalid audio fragment, and delete the invalid audio fragment.
8. the method for audio processing according to claim 6, which is characterized in that according to the control information, to described
After target audio carries out editing processing, the method for the audio processing further include:
The affiliated theme of first section audio fragment is obtained, and as the target topic；
The target audio in addition to the first section audio fragment is split as N number of audio fragment, N is the integer greater than 1, and right
Each audio fragment is handled as follows:
Semantics recognition is carried out to the audio fragment, obtains the semantics recognition result of the audio fragment；
The affiliated theme of the audio fragment is determined according to the semantics recognition result of the audio fragment；
The affiliated theme of the audio fragment is compared with the target topic, however, it is determined that the affiliated theme of audio fragment with
The target topic is not identical, then deletes the audio fragment.
9. a kind of electronic equipment characterized by comprising
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
It manages device to execute, so that at least one described processor is able to carry out the side of recording audio as described in any one in claim 1-5
Method.
10. a kind of server-side characterized by comprising
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
It manages device to execute, so that at least one described processor is able to carry out the side such as the described in any item audio processings of claim 6-8
Method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910147012.XA CN110111816B (en) | 2019-02-27 | 2019-02-27 | Audio recording method, audio processing method, electronic equipment and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910147012.XA CN110111816B (en) | 2019-02-27 | 2019-02-27 | Audio recording method, audio processing method, electronic equipment and server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110111816A true CN110111816A (en) | 2019-08-09 |
| CN110111816B CN110111816B (en) | 2021-03-05 |
Family
ID=67484251
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910147012.XA Active CN110111816B (en) | 2019-02-27 | 2019-02-27 | Audio recording method, audio processing method, electronic equipment and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110111816B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113613068A (en) * | 2021-08-03 | 2021-11-05 | 北京字跳网络技术有限公司 | Video processing method and device, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102568473A (en) * | 2011-12-30 | 2012-07-11 | 深圳市车音网科技有限公司 | Method and device for recording voice signals |
| CN104038630A (en) * | 2014-05-28 | 2014-09-10 | 小米科技有限责任公司 | Speech processing method and device |
| CN104869233A (en) * | 2015-04-27 | 2015-08-26 | 深圳市金立通信设备有限公司 | Recording method |
| CN104952451A (en) * | 2015-06-08 | 2015-09-30 | 广东欧珀移动通信有限公司 | Sound recording processing method and sound recording processing device |
| CN107071575A (en) * | 2016-06-13 | 2017-08-18 | 腾讯科技（北京）有限公司 | Paster media file playing method and device |
| CN107066229A (en) * | 2017-01-24 | 2017-08-18 | 广东欧珀移动通信有限公司 | The method and terminal of recording |
| CN107464557A (en) * | 2017-09-11 | 2017-12-12 | 广东欧珀移动通信有限公司 | Call recording method, device, mobile terminal and storage medium |
-
2019
- 2019-02-27 CN CN201910147012.XA patent/CN110111816B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102568473A (en) * | 2011-12-30 | 2012-07-11 | 深圳市车音网科技有限公司 | Method and device for recording voice signals |
| CN104038630A (en) * | 2014-05-28 | 2014-09-10 | 小米科技有限责任公司 | Speech processing method and device |
| CN104869233A (en) * | 2015-04-27 | 2015-08-26 | 深圳市金立通信设备有限公司 | Recording method |
| CN104952451A (en) * | 2015-06-08 | 2015-09-30 | 广东欧珀移动通信有限公司 | Sound recording processing method and sound recording processing device |
| CN107071575A (en) * | 2016-06-13 | 2017-08-18 | 腾讯科技（北京）有限公司 | Paster media file playing method and device |
| CN107066229A (en) * | 2017-01-24 | 2017-08-18 | 广东欧珀移动通信有限公司 | The method and terminal of recording |
| CN107464557A (en) * | 2017-09-11 | 2017-12-12 | 广东欧珀移动通信有限公司 | Call recording method, device, mobile terminal and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113613068A (en) * | 2021-08-03 | 2021-11-05 | 北京字跳网络技术有限公司 | Video processing method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110111816B (en) | 2021-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109309751B (en) | Voice recording method, electronic device and storage medium | |
| CN104394126B (en) | Information recommendation method, server, client and system | |
| CN107147618A (en) | A kind of user registering method, device and electronic equipment | |
| CN105554027A (en) | Resource sharing method and device | |
| JP5271703B2 (en) | Context sensitive data processing methods | |
| CN109284142A (en) | File preloads method, apparatus, electronic equipment and computer readable storage medium | |
| CN105187733A (en) | Video processing method, device and terminal | |
| CN107609047A (en) | Using recommendation method, apparatus, mobile device and storage medium | |
| CN108271096A (en) | A kind of task executing method, device, intelligent sound box and storage medium | |
| CN106507184A (en) | Media file shares terminal, receiving terminal, transmission method and electronic equipment | |
| CN109599115A (en) | Minutes method and apparatus for audio collecting device and user terminal | |
| CN107831886A (en) | Association starts management-control method, device, storage medium and the intelligent terminal of application | |
| CN111813900A (en) | Multi-turn conversation processing method and device, electronic equipment and storage medium | |
| CN106603649A (en) | Terminal equipment, booking event prompt method and apparatus thereof | |
| CN107862203A (en) | Control method, device, storage medium and the terminal of application program | |
| WO2001059607A3 (en) | Entertainment file and related information integration method, apparatus and system | |
| CN110111816A (en) | Method, the method for audio processing, electronic equipment and the server-side of recording audio | |
| CN110278273A (en) | Multimedia file method for uploading, device, terminal, server and storage medium | |
| CN110472033A (en) | Answering method, device and server based on NLP model | |
| CN111833857A (en) | Voice processing method and device and distributed system | |
| CN113949739B (en) | Cross-device playing method and device, electronic device and storage medium | |
| CN113163255B (en) | Video playing method, device, terminal and storage medium | |
| CN106708582A (en) | Data storage method and device | |
| CN108848472A (en) | The method and device of change of voice call | |
| CN108762633A (en) | Picture adding method, device, terminal device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |