Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Audio classification method and apparatus thereof

A classification method and audio technology, applied in speech analysis, speech recognition, instruments, etc., can solve the problem of low audio classification accuracy

Inactive Publication Date: 2016-07-20
LETV HLDG BEIJING CO LTD +1
View PDF4 Cites 34 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Embodiments of the present invention provide an audio classification method and device to solve the problem of low audio classification accuracy in the prior art

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
  • Audio classification method and apparatus thereof
  • Audio classification method and apparatus thereof
  • Audio classification method and apparatus thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0025] refer to figure 1 , which shows a flow chart of the steps of Embodiment 1 of an audio classification method of the present invention, which may specifically include:

[0026] Step 101. According to the collected training data, an audio classification model is obtained based on deep neural network training;

[0027] Step 102, extracting audio features to audio data;

[0028] Step 103: Input the audio feature into the audio classification model, and output the classification result of the audio data; the classification result includes: recording audio, voice search song audio and humming audio.

[0029] The embodiment of the present invention can be used to classify audio data through a smart terminal. Specifically, firstly, an audio classification model based on a deep neural network can be trained on the smart terminal according to the collected training data, then the audio features are extracted from the audio data to be classified, and finally the extracted audio f...

Embodiment 2

[0038] On the basis of the first embodiment above, this embodiment describes the specific process of extracting the audio features corresponding to the audio data; For the fundamental frequency features of humming audio and voice search audio with smaller frames, this embodiment further dynamically expands the extracted audio features by calculating the first-order and second-order difference operations, so that the audio features are more prominent, and finally combined into 42 Dimensional audio features.

[0039] refer to figure 2 , which shows a flow chart of the steps of an embodiment of a method for extracting audio features from audio data according to the present invention, which may specifically include:

[0040] Step 201, extracting the Mel cepstral coefficient feature and fundamental frequency feature corresponding to the audio data / training data;

[0041] In order to solve the problem of low accuracy in classifying voice search song audio and humming audio in the p...

Embodiment 3

[0068] This embodiment describes the training process of the audio classification model on the basis of the first embodiment above. refer to image 3 , which shows a flow chart of the steps of an embodiment of a training method for an audio classification model of the present invention, which may specifically include:

[0069] Step 301, collecting training data; the training data may specifically include: recording audio, voice search song audio and humming audio;

[0070] In the embodiment of the present invention, the collected training data may specifically include recording audio, voice search song audio, and humming audio, so as to meet common audio classification requirements. In order to enable the trained classification model to better avoid the interference caused by noise and silence, the embodiment of the present invention can also collect noise audio and mute audio as training data, and further, can also split the humming audio For singing audio and humming audio...

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

No PUM Login to View More

Abstract

Embodiments of the invention provide an audio classification method and an apparatus thereof. The method comprises the following steps of according to collected training data, based on a depth nerve network, training and acquiring an audio classification model; extracting an audio characteristic from the audio data; and inputting the audio characteristic into the audio classification model, outputting and acquiring a classification result of the audio data, wherein the classification result includes a recording audio, a voice song searching audio and a humming audio. In the prior art, a classification correct rate between the humming audio and the voice song searching audio is low. In the embodiments of the invention, the above problem can be solved, audio classification accuracy is increased and then song searching accuracy can be increased too.

Description

technical field [0001] The embodiments of the present invention relate to the field of audio technology, and in particular, to an audio classification method and device. Background technique [0002] In recent years, with the rapid development of the smart TV technology, more and more functions can be realized through the smart TV, for example, the function of searching songs can be realized through the smart TV. [0003] In specific applications, smart TVs can support the song search function in the following three ways: the first one is to receive the user's voice search song audio, for example, the voice search song audio can be a passage spoken by the received user: "Search song "Blue and White Porcelain", the smart TV will search in the search engine corresponding to the voice search audio; the second is to receive a piece of recorded audio input by the user, such as a recorded piece of background music, and the smart TV will search in the corresponding to the recorded ...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/02G10L15/06G10L15/16
CPCG10L15/02G10L15/063G10L15/16
Inventor 张利
Owner LETV HLDG BEIJING CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com