CN105512272A - System for comparing audio information and audio information comparison method - Google Patents

System for comparing audio information and audio information comparison method Download PDF

Info

Publication number
CN105512272A
CN105512272A CN201510883329.1A CN201510883329A CN105512272A CN 105512272 A CN105512272 A CN 105512272A CN 201510883329 A CN201510883329 A CN 201510883329A CN 105512272 A CN105512272 A CN 105512272A
Authority
CN
China
Prior art keywords
sample
file
wav
audio
comparison
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
Application number
CN201510883329.1A
Other languages
Chinese (zh)
Inventor
张明玉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tianjin Linkhope Technology Co ltd
Original Assignee
Tianjin Linkhope Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Tianjin Linkhope Technology Co ltd filed Critical Tianjin Linkhope Technology Co ltd
Priority to CN201510883329.1A priority Critical patent/CN105512272A/en
Publication of CN105512272A publication Critical patent/CN105512272A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/68Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/683Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content

Landscapes

  • Engineering & Computer Science (AREA)
  • Library & Information Science (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

A system for comparing audio information and an audio information comparison method relate to a system and a comparison method, in particular to a system for comparing audio information and an audio information comparison method. The invention aims to solve the problem that no method can be used for further manually verifying the sample retrieval matching result at present. The audio sample retrieval comparison module is used for carrying out position retrieval on target fragment audio and sample audio data through a sample retrieval method to obtain a matching position and similarity, and the comparison result processing and interface display module is used for further processing a matching result returned by the audio sample retrieval comparison module and displaying the matching result on an interface in a waveform form. The invention compares the target audio file with the sample file.

Description

A kind of system for comparison audio-frequency information and audio-frequency information comparison method
Technical field
The present invention relates to a kind of system and comparison method, be specifically related to a kind of system for comparison audio-frequency information and audio-frequency information comparison method, belong to computer software fields.
Background technology
Along with the arrival of cybertimes, network is full of the multimedia messages of magnanimity, automatically detect abroad to a large amount of flames of domestic propagation in order to effective, content-based network multimedia information on-line analysis system is developed and drops into application, thus broken away from the trouble of manual detection, achieve good Detection results, the while that on-line analysis system general effect being obvious, also there is a small amount of detection error situation, how further manual verification is carried out to sample retrieval matching result and become problem demanding prompt solution.
Summary of the invention
The present invention solves not have method can retrieve to sample the problem that matching result carries out further manual verification at present, and then proposes a kind of system for comparison audio-frequency information and audio-frequency information comparison method.
The present invention is the technical scheme taked that solves the problem: the present invention includes sample file layout conversion operations module, multimedia wav formatted file is play and control module, audio frequency sample retrieval contrast module and comparison result process and interface display module, it is wav formatted file that sample file layout conversion operations module is used for certain the sample file transform in sample storehouse, multimedia wav formatted file is play and control module has been used for being play wav formatted file by directshow, and control playing process with user interactions, audio frequency sample retrieval contrast module is used for target fragment audio frequency and sample voice data to carry out location retrieval by sample search method, obtain matched position and similarity, the matching result that comparison result process and interface display module are used for audio frequency sample retrieval comparing module returns is further processed, and matching result display is presented on interface with the form of waveform.
The concrete steps of comparison method of the present invention are as follows:
Step one, be mod formatted file by the sample document definition in sample library;
Step 2, in the sample file being defined as mod form, extract wav function, the sample file transform being defined as mod form is the sample file of wav form and preserves;
Step 3, in the sample file being converted to wav form the sample file of the retrieval wav form similar to the file destination of wav form, concrete operation step is:
Step 3 (one), the file destination of wav form is divided into multiple fragment, namely with reference to fragment, obtains each feature with reference to fragment from the statistical value of audio frame feature;
Step 3 (two), to compare each with the sample file of wav form successively with reference to fragment, calculate coupling similar value and matched position based on measuring similarity function;
Step 3 (three), similar value and a threshold value preset to be compared, threshold value is greater than 50%, if described similar value is greater than threshold value, then think that the sample file of this reference fragment and wav form has matched, and then the match information of the file destination of wav form and the sample file of wav form can be obtained, if similar value is less than threshold value, then return step 3 (two).
Step 4, the matching result that returns of audio frequency sample retrieval comparison to be processed, and the sample file of the file destination of wav form with the wav form mated is presented on interface with the form of waveform.
The invention has the beneficial effects as follows: the needs that present invention accomplishes audio frequency sample Compare System, utilize sample searching system to provide the method retrieving specific voice data in primary data source, the invention provides the verification method of sample retrieval matching result in on-line analysis system.It adopts segmentation-Based Audio Retrieval Algorithm, the audio fragment with standard audio storehouse sound intermediate frequency sample with certain similarity degree is detected from audio stream to be measured, comprise: searched targets is divided into multiple less fragment, namely with reference to fragment, each feature with reference to fragment is obtained from the statistical value of audio frame feature; Compare each with the audio frequency sample in standard audio storehouse successively with reference to fragment, calculate coupling similar value and matched position based on measuring similarity function; Similar value and a threshold value preset are compared, if described similar value is greater than threshold value, then thinks that the audio frequency sample in this reference fragment and standard sample storehouse has matched, and then the match information in searched targets and standard audio storehouse can be obtained.Searched targets segmentation is carried out independent retrieval by the present invention, and retrieval rate by searched targets effect length, is not applicable to real-time application scenario and retrieves specific voice data from unknown data source.
Accompanying drawing explanation
Fig. 1 is audio frequency sample Compare System process flow diagram of the present invention, and Fig. 2 is audio frequency sample Compare System interface display figure of the present invention.
Embodiment
Embodiment one: composition graphs 1 illustrates present embodiment, a kind of system for comparison audio-frequency information described in present embodiment comprises sample file layout conversion operations module 1, multimedia wav formatted file is play and control module 2, audio frequency sample retrieval contrast module 3 and comparison result process and interface display module 4, sample file layout conversion operations module 1 is for being wav formatted file by certain the sample file transform in sample storehouse, multimedia wav formatted file is play and control module 2 is play wav formatted file by directshow for completing, and control playing process with user interactions, audio frequency sample retrieval contrast module 3 is for carrying out location retrieval by target fragment audio frequency and sample voice data by sample search method, obtain matched position and similarity, comparison result process and interface display module 4 are further processed for the matching result returned audio frequency sample retrieval comparing module, and matching result display is presented on interface with the form of waveform.
Embodiment two: composition graphs 1 and Fig. 2 illustrate present embodiment, the concrete steps stating a kind of audio-frequency information comparison method described in present embodiment are as follows:
Step one, be mod formatted file by the sample document definition in sample library;
Step 2, in the sample file being defined as mod form, extract wav function, the sample file transform being defined as mod form is the sample file of wav form and preserves;
Step 3, in the sample file being converted to wav form the sample file of the retrieval wav form similar to the file destination of wav form, concrete operation step is:
Step 3 (one), the file destination of wav form is divided into multiple fragment, namely with reference to fragment, obtains each feature with reference to fragment from the statistical value of audio frame feature;
Step 3 (two), to compare each with the sample file of wav form successively with reference to fragment, calculate coupling similar value and matched position based on measuring similarity function;
Step 3 (three), similar value and a threshold value preset to be compared, threshold value is greater than 50%, if described similar value is greater than threshold value, then think that the sample file of this reference fragment and wav form has matched, and then the match information of the file destination of wav form and the sample file of wav form can be obtained, if similar value is less than threshold value, then return step 3 (two).
Step 4, the matching result that returns of audio frequency sample retrieval comparison to be processed, and the sample file of the file destination of wav form with the wav form mated is presented on interface with the form of waveform.
Each in step 3 () in present embodiment is 3s with reference to clip size, from the sample file of wav form, intercept according to retrieving the matching value position returned the audio section comprising the common 9s of the file destination audio frequency of wav form simultaneously, the reference fragment of 3s presents with waveform at interface with the audio frequency that mates of 9s simultaneously, whether correctly distinguishes for user.

Claims (2)

1. the system for comparison audio-frequency information, it is characterized in that: described a kind of system for comparison audio-frequency information comprises sample file layout conversion operations module (1), multimedia wav formatted file is play and control module (2), audio frequency sample retrieval contrast module (3) and comparison result process and interface display module (4), sample file layout conversion operations module (1) is for being wav formatted file by certain the sample file transform in sample storehouse, multimedia wav formatted file is play and control module (2) is play wav formatted file by directshow for completing, and control playing process with user interactions, audio frequency sample retrieval contrast module (3) is for carrying out location retrieval by target fragment audio frequency and sample voice data by sample search method, obtain matched position and similarity, comparison result process and interface display module (4) are further processed for the matching result returned audio frequency sample retrieval comparing module, and matching result display is presented on interface with the form of waveform.
2. utilize system described in claim 1 to carry out a method for audio-frequency information comparison, it is characterized in that: the concrete steps of described a kind of audio-frequency information comparison method are as follows:
Step one, be mod formatted file by the sample document definition in sample library;
Step 2, in the sample file being defined as mod form, extract wav function, the sample file transform being defined as mod form is the sample file of wav form and preserves;
Step 3, in the sample file being converted to wav form the sample file of the retrieval wav form similar to the file destination of wav form, concrete operation step is:
Step 3 (one), the file destination of wav form is divided into multiple fragment, namely with reference to fragment, obtains each feature with reference to fragment from the statistical value of audio frame feature;
Step 3 (two), to compare each with the sample file of wav form successively with reference to fragment, calculate coupling similar value and matched position based on measuring similarity function;
Step 3 (three), similar value and a threshold value preset are compared, threshold value is greater than 50%, if described similar value is greater than threshold value, then think that the sample file of this reference fragment and wav form has matched, and then the match information of the file destination of wav form and the sample file of wav form can be obtained, if similar value is less than threshold value, then return step 3 (two), step 4, the matching result that audio frequency sample retrieval comparison returns is processed, and the sample file of the file destination of wav form with the wav form mated is presented on interface with the form of waveform.
CN201510883329.1A 2015-12-04 2015-12-04 System for comparing audio information and audio information comparison method Pending CN105512272A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510883329.1A CN105512272A (en) 2015-12-04 2015-12-04 System for comparing audio information and audio information comparison method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510883329.1A CN105512272A (en) 2015-12-04 2015-12-04 System for comparing audio information and audio information comparison method

Publications (1)

Publication Number Publication Date
CN105512272A true CN105512272A (en) 2016-04-20

Family

ID=55720254

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510883329.1A Pending CN105512272A (en) 2015-12-04 2015-12-04 System for comparing audio information and audio information comparison method

Country Status (1)

Country Link
CN (1) CN105512272A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107516534A (en) * 2017-08-31 2017-12-26 广东小天才科技有限公司 A kind of comparison method of voice messaging, device and terminal device
CN112562732A (en) * 2020-12-24 2021-03-26 北京睿芯高通量科技有限公司 Audio analysis system and analysis method thereof
CN113885828A (en) * 2021-10-25 2022-01-04 北京字跳网络技术有限公司 Sound effect display method and terminal equipment

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107516534A (en) * 2017-08-31 2017-12-26 广东小天才科技有限公司 A kind of comparison method of voice messaging, device and terminal device
CN107516534B (en) * 2017-08-31 2020-11-03 广东小天才科技有限公司 Voice information comparison method and device and terminal equipment
CN112562732A (en) * 2020-12-24 2021-03-26 北京睿芯高通量科技有限公司 Audio analysis system and analysis method thereof
CN112562732B (en) * 2020-12-24 2024-04-16 北京中科通量科技有限公司 Audio analysis system and analysis method thereof
CN113885828A (en) * 2021-10-25 2022-01-04 北京字跳网络技术有限公司 Sound effect display method and terminal equipment
CN113885828B (en) * 2021-10-25 2024-03-12 北京字跳网络技术有限公司 Sound effect display method and terminal equipment

Similar Documents

Publication Publication Date Title
US11064227B2 (en) Systems and methods for live media content matching
US10657325B2 (en) Method for parsing query based on artificial intelligence and computer device
US10133538B2 (en) Semi-supervised speaker diarization
CN101159834B (en) Method and system for detecting repeatable video and audio program fragment
US9756368B2 (en) Methods and apparatus to identify media using hash keys
US11665288B2 (en) Methods and apparatus to identify media using hybrid hash keys
WO2019196205A1 (en) Foreign language teaching evaluation information generating method and apparatus
CN101221760B (en) Audio matching method and system
CN102799605A (en) Method and system for monitoring advertisement broadcast
JP2013525916A5 (en)
US10943600B2 (en) Systems and methods for interrelating text transcript information with video and/or audio information
JP2019212292A (en) Event detection method, device, equipment, and program
CN105512272A (en) System for comparing audio information and audio information comparison method
CN105632487A (en) Voice recognition method and device
CN103605666A (en) Video copying detection method for advertisement detection
KR102492049B1 (en) Media identification using watermarks and signatures
CN110674638A (en) Corpus labeling system and electronic equipment
CN113194332B (en) Multi-policy-based new advertisement discovery method, electronic device and readable storage medium
CN102855473B (en) A kind of image multi-target detection method based on similarity measurement
CN103077203A (en) Method for detecting repetitive audio/video clips
CN107133644B (en) Digital library's content analysis system and method
WO2017107651A1 (en) Method and device for determining relevance between news and for calculating the relevance between news
Stein et al. From raw data to semantically enriched hyperlinking: Recent advances in the LinkedTV analysis workflow
CN113923479A (en) Audio and video editing method and device
CN113569096B (en) Structured information extraction method, device, equipment and storage medium

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20160420