Audio frequency rapid classification method based on content
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A rapid classification and audio technology, applied in the field of information processing, can solve the problems of complex classification, impossibility, and small number of classifications, and achieve high efficiency
Inactive Publication Date: 2009-06-03
SHENZHEN RAISOUND TECH +1
View PDF0 Cites 40 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0006] 2. The category of classification cannot be flexibly changed according to requirements
Therefore, when the requirements change, the whole system needs to be greatly adjusted, and the performance of the classification cannot be guaranteed
[0007] 3. Complex classification cannot be realized
The above-mentioned audio classification method has a small number of classifications, and at the same time cannot achieve complex classifications, such as requiring a certain word to appear in the speech, a certain word does not appear in the speech, or classification of logical combinations such as two words appearing at the same time
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment approach 1
[0077] Embodiment 1: Convert the speech feature vector of the speech vector segment into a speech content sequence represented by basic acoustic model units (AU) according to the acoustic model, language model and dictionary.
[0078] Speech recognition of speech vector segments requires the use of acoustic models, language models, and dictionaries.
[0079] (1) Acoustic model (AM): The acoustic model includes several basic acoustic model units (AU). AU is the representation of the speech feature vector corresponding to the pronunciation unit (phonetic symbol) of each basic speech, which is complete and distinguishable.
[0080] Completeness: All possible pronunciation units (phonetic symbols) in speech have their corresponding speech feature vector representations.
[0081] Distinction: Different pronunciation units (phonetic symbols) should not be exactly the same.
[0082] Wherein, each pronunciation unit (phonetic symbol) AU corresponds to a segment of hundreds of voice f...
Embodiment approach 2
[0087] Embodiment 2: Convert the speech feature vector of the speech vector segment into a speech content sequence represented by basic acoustic model units (AU) according to the acoustic model and the dictionary.
[0088] Speech recognition of speech vector segments requires the use of acoustic models and dictionaries.
[0089] (1) Acoustic model (AM): The acoustic model includes several basic acoustic model units (AU). AU is the representation of the speech feature vector corresponding to the pronunciation unit (phonetic symbol) of each basic speech, which is complete and distinguishable.
[0090] Completeness: All possible pronunciation units (phonetic symbols) in speech have their corresponding speech feature vector representations.
[0091] Distinction: Different pronunciation units (phonetic symbols) should not be exactly the same.
[0092] Wherein, each pronunciation unit (phonetic symbol) AU corresponds to a segment of hundreds of voice feature vector sequences in the...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The invention discloses an audio frequency rapid classification method based on content, comprising the following steps: pretreatment and character extraction are carried out on audio frequency data, and a series of speech feature vectors are obtained; the speech feature vectors are converted into speech content sequences which are represented by a basic acoustic model unit AU and the speech content sequences comprising classification key words are recognized according to a created word list; the confidence score of every speech content sequence comprising the classification key words is calculated; whether the confidence score is within a set threshold distribution range is judged, and if yes, recognizing results of the speech content sequences comprising the classification key words are reserved; the recognizing results are counted, and the audio frequency data which accord with set conditions are classified according to the set of a classification task on the classification key words. The method fully utilizes content information and can realize complex classifications conveniently and process a plurality of audio data files in a parallel way. The efficiency of the file classification processing is high.
Description
【Technical field】 [0001] The invention relates to the technical field of information processing, in particular to a content-based audio rapid classification method. 【Background technique】 [0002] With the rapid development of modern society, there are more and more text, image, video and audio information. In the field of communication and Internet, audio information occupies a very important position. In the various processing of audio information, audio classification is one of the very important processing processes. [0003] A current technical scheme of an audio classification method is to first preprocess the input audio signal, then calculate the linear predictive coding coefficient of the audio signal, and then obtain the spectral envelope of the signal according to the linear predictive coding coefficient, and then calculate the obtained The guide spectrum determines the amplitude difference value of the parameters, and finally classifies the audio signal accordi...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more