CN110310661A - A kind of calculation method of two-way real-time broadcast audio delay and similarity - Google Patents
A kind of calculation method of two-way real-time broadcast audio delay and similarity Download PDFInfo
- Publication number
- CN110310661A CN110310661A CN201910592528.5A CN201910592528A CN110310661A CN 110310661 A CN110310661 A CN 110310661A CN 201910592528 A CN201910592528 A CN 201910592528A CN 110310661 A CN110310661 A CN 110310661A
- Authority
- CN
- China
- Prior art keywords
- audio
- sampling data
- way
- similarity
- delay
- 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
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
- G10L25/21—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being power information
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
- G10L25/24—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being the cepstrum
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04H—BROADCAST COMMUNICATION
- H04H20/00—Arrangements for broadcast or for distribution combined with broadcast
- H04H20/28—Arrangements for simultaneous broadcast of plural pieces of information
- H04H20/33—Arrangements for simultaneous broadcast of plural pieces of information by plural channels
Abstract
The present invention relates to the calculation methods of a kind of two-way real-time broadcast audio delay and similarity, belong to real time audio processing technical field.Obtain the real-time broadcast audio flow data of two-way that length is X secondsBy audio sampling data matrixPostpone to obtain audio sampling data matrix in m seconds backwardAgain by audio sampling data matrixWith audio sampling data matrixIt is superimposed to obtain the audio sampling data of mixed signal according to specified ruleCalculate mixed signalPower cepstrum the delay between two-way audio is calculated by the position at the top of the first half in power cepstrum matrix, the similarity percentage between two-way audio is calculated by the amplitude at the top found out by k number preceding in power cepstrum matrix according to being set to 0.Calculating speed of the present invention is very fast and fault-tolerance with higher, also delay product and similarity result can be made more accurate when broadcast audio is by noise jamming.
Description
Technical field
The present invention relates to the calculation methods of a kind of two-way real-time broadcast audio delay and similarity, belong to real time audio processing
Technical field.
Background technique
In broadcast transmitting station, for the input signal of broadcast audio typically from satellite transmission, which passes through transmitter
Modulation is emitted with medium wave or short frequency, in order to guarantee that the audio signal transmitted during the launch process is not tampered or pollutes,
The reception demodulation result to the transferred signal of transmitter and radio station transmitting signal is needed to be compared.This course of work is big at present
It is manually carried out more, i.e., transmitting signal is received by radio, sound distinguishing by ear.Broadcast transmitting station often has more
Road audio input signal needs to emit, so the workload that artificial audio prison is broadcast is very big, is unable to reach the requirement supervised broadcast in real time.
However, the audio signal of same programs is typically all to be passed to by different transmission links.Broadcast voice signal
After different links (satellite, optical fiber, microwave, network, modulation, demodulation) transmission, the time that they reach program monitoring point is
There are time delays.And audio comparison monitoring is carried out in the state that two-way audio signal is asynchronous cannot obtain accurate phase
Like degree result.Therefore the delay between estimation two-way audio signal is the premise and item of audio comparison monitoring to make it synchronize
Part, the accuracy of synchronous result meeting extreme influence audio comparison result.
In general, traditional audio similarity calculation method first extracts the characteristic parameter of audio, then the difference of comparative feature parameter
Similarity that is different and obtaining two-way audio, this method can only be directed to ordinary audio, not be suitable for broadcast audio, reason is main
In terms of having following three: in a first aspect, the similarity result that this method obtains is just if there is delay between two-way real-time audio
It is low.Second aspect, on comparing fault-tolerance, numerous audio frequency characteristics are pursued high content and are distinguished all using accurate description audio as target
Knowledge and magnanimity, therefore the slight change of audio can react characteristically.In this way, if audio is some often by noise jamming, or process
Filter process, such as bandpass filter are advised, then treated audio is compared with original audio, traditional similarity calculating method
It is dissimilar or even entirely different often to judge two audios, and noise jamming, filtering processing etc. are very common in the broadcast.Such as exist
In broadcast audio monitoring, we need the audio of broadcast and the audio received progress similarity calculation, and the sound received
Frequency usually contains noise；Subjective audibility is preferably for another example broadcasted in order to reach, it is equal usually using audio before broadcast transmission
Weighing apparatus come to broadcast audio frequency spectrum be adjusted.These processing so that audio processed has occurred larger relative to former audio
Variation.In this case, traditional similarity calculating method tends not to be applicable in, because broadcast audio comparison needs to determine
Whether whether audio content of broadcast itself is identical, rather than consistent on auditory effect, even if that is, certain audio is by certain noise
Interference, filter process, balanced device adjust etc. processing after, algorithm should still determine the audio and its original audio be similar.The
Three aspects extract audio frequency characteristics parameter in the computational efficiency of similarity to calculate the method meter of two-way real-time audio similarity
Calculation complexity is relatively high, and the required time is relatively long, is unfavorable for handling in real time.
Summary of the invention
The technical problem to be solved by the present invention is to no longer use traditional for the Problems Existing being previously mentioned in background technique
The mode of audio frequency characteristics parameter extraction calculates two-way real-time audio similarity, and is to provide a kind of two-way real-time broadcast audio and prolongs
When and similarity calculation method.
The technical scheme is that a kind of calculation method of two-way real-time broadcast audio delay and similarity, obtains length
The real-time broadcast audio flow data of two-way that degree is X secondsBy audio sampling data matrixPostpone to obtain for m seconds backward
Audio sampling data matrixAgain by audio sampling data matrixWith audio sampling data matrixIt is superimposed according to specified rule
Obtain the audio sampling data of mixed signalCalculate mixed signalPower cepstrum, by preceding k in power cepstrum matrix
A data are set to 0, calculate the delay between two-way audio by the position at the top of the first half in power cepstrum matrix, by
The amplitude at the top found out calculates the similarity percentage between two-way audio.
Specific steps are as follows:
Step1: the real-time broadcast audio stream data matrix of two-way that length is X seconds is obtainedTo first via audio
Signal and the second tunnel audio signal, timing acquiring X seconds audio respectively, to two section audios according to specified sample frequency f into
Row sampling, obtains two groups of audio sampling data matrixesThere is N number of sample point data in each matrix, i.e.,
Step2: judge whether two-way audio signal is mute state: according to two groups of audio sampling data matrixes
In the level values of all sampled points judge whether two-way audio signal is mute state respectively；If wherein audio is mute shape all the way
State then continues to execute Step1 and Step2, if following the steps below if two-way audio two-way audio is not static state；
Step3: by audio sampling data matrixPostpone to obtain audio sampling data matrix in m seconds backward
Step4: by audio sampling data matrixWith audio sampling data matrixIt is superimposed and is mixed according to specified rule
Close the audio sampling data of signalSpecified rule are as follows:
Step5: mixed signal is calculatedPower cepstrum；
Step6: k number evidence preceding in power cepstrum matrix is set to 0；The value of the k can be strained mutually according to the actual situation
Change, generally takes 200 to 400.
Step7: finding out top from the first half of power cepstrum matrix, and judges whether peak value is greater than threshold value, if greatly
In threshold value, then Step8 is carried out, determines that two audios are completely dissimilar if being less than threshold value, similarity 0；
Step8: calculating the delay between two-way audio by the position at the top found out, by the width at the top found out
Degree calculates the similarity percentage between two-way audio according to specified rule.The threshold value can be strained mutually according to the actual situation
Change, generally takes 0.02 to 0.03.
According to two groups of audio sampling data matrixes in the Step2In the level values of all sampled points sentence respectively
The step of whether disconnected two-way audio signal is mute state are as follows: respectively to two groups of audio sampling data matrixesIt is all
Sampled point level value seeks absolute value and sums, it may be assumed that
In formula, aiFor audio sampling data matrixIn i-th of audio sampling data, biFor audio sampling data matrix
In i-th of audio sampling data；
It is judged as mute state if being less than threshold value, is judged as non-mute state if more than threshold value.The threshold value can root
Factually border concrete condition and change.
By audio sampling data matrix in the Step3Postpone to obtain audio sampling data matrix in m seconds backwardTool
Body step are as follows: first by audio sampling data matrixIn preceding mf element be assigned a value of 0, then by audio sampling data matrix
In preceding N-mf element successively copy to audio sampling data matrixIn.Delay m seconds of addition can become according to the actual situation
Change, generally takes 0.1 second to 0.3 second, it is unsuitable too long.
Mixed signal is calculated in the Step4Power cepstrum are as follows:
Step5.1: mixed signal is calculatedFast Fourier Transform and modulus；
Step5.2: the logarithm using e the bottom of as is asked to Step5.1 acquired results；
Step5.3: fast adverse Fourier transform and modulus are asked to Step5.2 acquired results.
The specified rule of the delay between two-way audio is calculated by the position at the top found out in the Step8 are as follows:
In formula, true delay of the Delay between two-way audio, peakpos is top in power cepstrum the first half
Position, m are delay m seconds added in Step3, and f is sample frequency；
The finger of the similarity percentage between two-way audio is calculated according to specified rule by the amplitude at the top found out
Set pattern is then are as follows:
In formula, similarity of the Similarity between two-way audio, peak is the amplitude at top.
The beneficial effects of the present invention are: real-time to calculate two-way no longer by the way of traditional audio frequency characteristics parameter extraction
Audio similarity, but the delay between two-way audio and two-way sound are calculated according to the feature of superposed signal power cepstrum peak
Similarity percentage between frequency can calculate similar between the delay and two-way audio between two-way audio in a short time
Spend percentage.The algorithm calculating speed is very fast and fault-tolerance with higher, can also make when broadcast audio is by noise jamming
It obtains delay product and similarity result is more accurate.
Detailed description of the invention
Fig. 1 is the waveform diagram of specific embodiment of the invention sound intermediate frequency chB1；
Fig. 2 is the waveform diagram of specific embodiment of the invention sound intermediate frequency chB2；
Fig. 3 is the spectrogram of specific embodiment of the invention sound intermediate frequency chB1；
Fig. 4 is the spectrogram of specific embodiment of the invention sound intermediate frequency chB2；
Fig. 5 is step flow chart of the invention；
Fig. 6 is the step flow chart that mixed signal power cepstrum is calculated in Step5 of the present invention.
Specific embodiment
With reference to the accompanying drawings and detailed description, the invention will be further described.
Emulation experiment has selected 6 audios as testing audio altogether, and audio data length is 30 seconds, is monophonic, adopts
Sample rate is all 8000Hz, and has done amplitude normalization processing.ChA1 and chA2 is same audio source, and chB1 and chB2 are another sounds
Source, i.e. chA1 is identical as chA2 audio content, and chB1 is identical as chB2 audio content.ChA1, chA2, chC1 are without band
The former broadcast audio of bandpass filter, chB1, chB2, chC2 are the audio obtained by the bandpass filter of different passbands, with mould
Intend different transmission channels.ChA2 ratio chA1 is delayed 2 seconds, and chB2 ratio chB1 is delayed 3 seconds, to simulate after different transmission channels
Generated delay between audio.The information of 6 audios is as shown in the table.
It is respectively combined by two audios of difference of above 6 audios, has carried out the emulation experiment of following 8 kinds of situations.
The present embodiment is to test chB1 the and chB2 two-way audio that the bandpass filter in 6 by different passbands is handled
For, it is compared, and calculate the delay and similarity between two-way audio, the specific steps are as follows:
Step1: the real-time broadcast audio stream data matrix of two-way that length is x seconds is obtainedTo first via audio
Signal and the second tunnel audio signal, timing acquires 8.192 seconds audios respectively, to two section audios according to specified sample frequency
8000Hz is sampled, and two groups of audio sampling data matrixes are obtainedThere are 65536 sample point datas in each matrix,
I.e.
Step2: judge whether two-way audio signal is mute state: according to two groups of audio sampling data matrixes
In the level values of all sampled points judge whether two-way audio signal is mute state respectively.Respectively to two groups of audio sampling datas
Matrix All sampled point level values seek absolute value and sum, it may be assumed that
It is 1044.128, Sum_B_sample 310.1565 that Sum_A_sample, which is calculated,.
Both it is greater than set threshold value (100), two-way audio two-way audio, can be into not for static state if then can determine whether
Row following steps.
Step3: by audio sampling data matrixPostpone to obtain audio sampling data matrix in 0.1 second backwardFirst by sound
Frequency sampling data matrixIn preceding 800 elements be assigned a value of 0, then by audio sampling data matrixIn preceding 64736 elements
Successively copy to audio sampling data matrixIn.
Step4: by audio sampling data matrixWith audio sampling data matrixAccording to the specified rule:
Superposition obtains the audio sampling data of mixed signal
Step5: mixed signal is calculatedPower cepstrum, and including following steps；
Step5.1: mixed signal is calculatedFast Fourier Transform and modulus；
Step5.2: the logarithm using e the bottom of as is asked to Step5.1 acquired results；
Step5.3: fast adverse Fourier transform and modulus are asked to Step5.2 acquired results.
Step6: 400 data preceding in power cepstrum matrix are set to 0.
Step7: it is 0.027947 that top is found from the first half of power cepstrum matrix, is greater than set threshold value 0.02,
It can carry out Step8.
Step8: the position at the top found out is 24441, according to specified rule:
In formula, true delay of the Delay between two-way audio, peakpos is top in power cepstrum the first half
Position, m are delay m seconds added in Step3, and f is sample frequency.
Calculate the delay between two-way audio are as follows: 2.955 seconds.
The amplitude at the top found out is 0.027947, according to specified rule:
Calculating the similarity percentage between two-way audio is 71.93%.
In conjunction with attached drawing, the embodiment of the present invention is explained in detail above, but the present invention is not limited to above-mentioned
Embodiment within the knowledge of a person skilled in the art can also be before not departing from present inventive concept
Put that various changes can be made.
Claims (5)
1. a kind of calculation method of two-way real-time broadcast audio delay and similarity, it is characterised in that:
Step1: the real-time broadcast audio stream data matrix of two-way that length is X seconds is obtainedTo first via audio signal
With the second tunnel audio signal, timing acquiring X seconds audio, two section audios are adopted according to specified sample frequency f respectively
Sample obtains two groups of audio sampling data matrixesThere is N number of sample point data in each matrix, i.e.,
Step2: judge whether two-way audio signal is mute state: according to two groups of audio sampling data matrixesMiddle institute
There is the level value of sampled point to judge whether two-way audio signal is mute state respectively；If wherein audio is mute state all the way,
Step1 and Step2 are then continued to execute, if following the steps below if two-way audio two-way audio is not static state；
Step3: by audio sampling data matrixPostpone to obtain audio sampling data matrix in m seconds backward
Step4: by audio sampling data matrixWith audio sampling data matrixIt is superimposed to obtain mixing letter according to specified rule
Number audio sampling dataSpecified rule are as follows:
Step5: mixed signal is calculatedPower cepstrum；
Step6: k number evidence preceding in power cepstrum matrix is set to 0；
Step7: finding out top from the first half of power cepstrum matrix, and judges whether peak value is greater than threshold value, if more than threshold
Value, then carry out Step8, determines that two audios are completely dissimilar if being less than threshold value, similarity 0；
Step8: calculating the delay between two-way audio by the position at the top found out, by the amplitude root at the top found out
The similarity percentage between two-way audio is calculated according to specified rule.
2. the calculation method of two-way real-time broadcast audio delay and similarity according to claim 1, it is characterised in that: institute
It states in Step2 according to two groups of audio sampling data matrixesIn the level values of all sampled points judge two-way audio respectively
The step of whether signal is mute state are as follows: respectively to two groups of audio sampling data matrixesAll sampling level points
Value seeks absolute value and sums, it may be assumed that
In formula, aiFor audio sampling data matrixIn i-th of audio sampling data, biFor audio sampling data matrixIn i-th
A audio sampling data；
It is judged as mute state if being less than threshold value, is judged as non-mute state if more than threshold value.
3. the calculation method of two-way real-time broadcast audio delay and similarity according to claim 1, it is characterised in that: institute
It states audio sampling data matrix in Step3Postpone to obtain audio sampling data matrix in m seconds backwardSpecific steps are as follows:
First by audio sampling data matrixIn preceding mf element be assigned a value of 0, then by audio sampling data matrixIn preceding N-
Mf element successively copies to audio sampling data matrixIn.
4. the calculation method of two-way real-time broadcast audio delay and similarity according to claim 1, it is characterised in that: institute
It states and calculates mixed signal in Step4Power cepstrum are as follows:
Step5.1: mixed signal is calculatedFast Fourier Transform and modulus；
Step5.2: the logarithm using e the bottom of as is asked to Step5.1 acquired results；
Step5.3: fast adverse Fourier transform and modulus are asked to Step5.2 acquired results.
5. the calculation method of two-way real-time broadcast audio delay and similarity according to claim 1, it is characterised in that: institute
It states in Step8 and to calculate the specified rule of the delay between two-way audio by the position at the top found out are as follows:
In formula, true delay of the Delay between two-way audio, peakpos is the position at top in power cepstrum the first half,
M is delay m seconds added in Step3, and f is sample frequency；
The specified rule of the similarity percentage between two-way audio are calculated according to specified rule by the amplitude at the top found out
Then are as follows:
In formula, similarity of the Similarity between two-way audio, peak is the amplitude at top.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910592528.5A CN110310661B (en) | 2019-07-03 | 2019-07-03 | Method for calculating two-path real-time broadcast audio time delay and similarity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910592528.5A CN110310661B (en) | 2019-07-03 | 2019-07-03 | Method for calculating two-path real-time broadcast audio time delay and similarity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110310661A true CN110310661A (en) | 2019-10-08 |
| CN110310661B CN110310661B (en) | 2021-06-11 |
Family
ID=68078816
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910592528.5A Active CN110310661B (en) | 2019-07-03 | 2019-07-03 | Method for calculating two-path real-time broadcast audio time delay and similarity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110310661B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112214635A (en) * | 2020-10-23 | 2021-01-12 | 昆明理工大学 | Fast audio retrieval method based on cepstrum analysis |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011065828A1 (en) * | 2009-11-27 | 2011-06-03 | Nederlandse Organisatie Voor Toegepast- Natuurwetenschappelijk Onderzoek Tno | Method for detecting audio ticks in a noisy environment |
| CN104505101A (en) * | 2014-12-24 | 2015-04-08 | 北京巴越赤石科技有限公司 | Real-time audio comparison method |
| CN104700842A (en) * | 2015-02-13 | 2015-06-10 | 广州市百果园网络科技有限公司 | Sound signal time delay estimation method and device |
| CN104900238A (en) * | 2015-05-14 | 2015-09-09 | 电子科技大学 | Audio real-time comparison method based on sensing filtering |
| CN104900239A (en) * | 2015-05-14 | 2015-09-09 | 电子科技大学 | Audio real-time comparison method based on Walsh-Hadamard transform |
| CN107274911A (en) * | 2017-05-03 | 2017-10-20 | 昆明理工大学 | A kind of similarity analysis method based on sound characteristic |
| CN107749300A (en) * | 2017-09-15 | 2018-03-02 | 苏州市福川科技有限公司 | Audio Compare System based on content |
| CN108665903A (en) * | 2018-05-11 | 2018-10-16 | 复旦大学 | A kind of automatic testing method and its system of audio signal similarity degree |
| US20180309421A1 (en) * | 2017-04-20 | 2018-10-25 | Dts, Inc. | Loudness control with noise detection and loudness drop detection |
-
2019
- 2019-07-03 CN CN201910592528.5A patent/CN110310661B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011065828A1 (en) * | 2009-11-27 | 2011-06-03 | Nederlandse Organisatie Voor Toegepast- Natuurwetenschappelijk Onderzoek Tno | Method for detecting audio ticks in a noisy environment |
| CN104505101A (en) * | 2014-12-24 | 2015-04-08 | 北京巴越赤石科技有限公司 | Real-time audio comparison method |
| CN104700842A (en) * | 2015-02-13 | 2015-06-10 | 广州市百果园网络科技有限公司 | Sound signal time delay estimation method and device |
| CN104900238A (en) * | 2015-05-14 | 2015-09-09 | 电子科技大学 | Audio real-time comparison method based on sensing filtering |
| CN104900239A (en) * | 2015-05-14 | 2015-09-09 | 电子科技大学 | Audio real-time comparison method based on Walsh-Hadamard transform |
| US20180309421A1 (en) * | 2017-04-20 | 2018-10-25 | Dts, Inc. | Loudness control with noise detection and loudness drop detection |
| CN107274911A (en) * | 2017-05-03 | 2017-10-20 | 昆明理工大学 | A kind of similarity analysis method based on sound characteristic |
| CN107749300A (en) * | 2017-09-15 | 2018-03-02 | 苏州市福川科技有限公司 | Audio Compare System based on content |
| CN108665903A (en) * | 2018-05-11 | 2018-10-16 | 复旦大学 | A kind of automatic testing method and its system of audio signal similarity degree |
Non-Patent Citations (4)
| Title |
|---|
| NANTAWAT JAIYEN: "Real-time audio similarity comparison algorithm", 《2017 IEEE 15TH STUDENT CONFERENCE ON RESEARCH AND DEVELOPMENT (SCORED)》 * |
| YUQING YE: "Audio similarity comparison system of english dubbing on android platform", 《2017 IEEE INTERNATIONAL CONFERENCE ON INFORMATION AND AUTOMATION (ICIA)》 * |
| 何连成: "音频信号内容比对实用算法", 《中国传媒科技》 * |
| 刘春江: "基于音频信号比对的无线广播节目智能监测系统", 《广播与电视技术》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112214635A (en) * | 2020-10-23 | 2021-01-12 | 昆明理工大学 | Fast audio retrieval method based on cepstrum analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110310661B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108899044B (en) | Voice signal processing method and device | |
| US10068586B2 (en) | Binaurally integrated cross-correlation auto-correlation mechanism | |
| US10360923B2 (en) | Method and system for eliminating an echo | |
| KR101670313B1 (en) | Signal separation system and method for selecting threshold to separate sound source | |
| EP3621072A1 (en) | Extracting and enhancing a watermark signal from an output signal of a watermarking encoder | |
| JPS63259696A (en) | Voice pre-processing method and apparatus | |
| CN104992713B (en) | A kind of quick broadcast audio comparison method | |
| CN109040911B (en) | Intelligent sound box and determination method for target placement position thereof | |
| CN107393555A (en) | A kind of detecting system and detection method of low signal-to-noise ratio abnormal sound signal | |
| JP2012226080A (en) | Music piece retrieval device and method, program, and recording medium | |
| US20210306782A1 (en) | Method and system of audio device performance testing | |
| CN110610718A (en) | Method and device for extracting expected sound source voice signal | |
| CN110310661A (en) | A kind of calculation method of two-way real-time broadcast audio delay and similarity | |
| US20080106249A1 (en) | Generating sample error coefficients | |
| US9866422B1 (en) | Methods of blind source separation filter resource management | |
| CN112053669B (en) | Method, device, equipment and medium for eliminating human voice | |
| CN104900239B (en) | A kind of audio real-time comparison method based on Walsh-Hadamard transform | |
| CN112992190B (en) | Audio signal processing method and device, electronic equipment and storage medium | |
| CN105429720A (en) | Related delay estimation method based on EMD reconstruction | |
| CN114927141B (en) | Method and system for detecting abnormal underwater acoustic signals | |
| CN115696176A (en) | Audio object-based sound reproduction method, device, equipment and storage medium | |
| Oliinyk et al. | Center weighted median filter application to time delay estimation in non-Gaussian noise environment | |
| CN110426711B (en) | Time delay estimation method and system based on polarity zero detection | |
| CN111028857B (en) | Method and system for reducing noise of multichannel audio-video conference based on deep learning | |
| WO2022219558A8 (en) | System and method for estimating direction of arrival and delays of early room reflections |
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 |