Unlock instant, AI-driven research and patent intelligence for your innovation.

Method and device for classifying audio data

A technology of audio data and classification methods, applied in the field of information processing, can solve problems such as poor classification processing effects, achieve good robustness, wide applicability, and improve adaptability

Active Publication Date: 2017-09-12
ZHEJIANG DAHUA TECH CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] An embodiment of the present invention provides a method for classifying audio data, which is used to solve the problem in the prior art that the classification processing effect is not good when processing audio data sequences with different speeds

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
  • Method and device for classifying audio data
  • Method and device for classifying audio data
  • Method and device for classifying audio data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0015] Through the analysis and research of the prior art, the inventor finds that the audio data classification method adopted in the prior art has a common defect, that is, the defect that the classification effect is not good when the audio data with different fast and slow rhythms are classified and processed: if the training If the fast audio data samples used in the training process account for the majority, but a piece of slow audio data of an unknown category is processed in the classification, the classification effect will be poor; if the slow audio data samples used in the training are in the majority, and If a piece of audio data of a relatively fast unknown category is processed during classification, the classification effect will be poor.

[0016] In order to provide a classification implementation scheme that can adapt to audio data with different speeds, an embodiment of the present invention provides a method for classifying audio data. The specific flow chart...

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

The invention discloses a method for audio data classification so as to solve the problem of poor classification processing effect in the prior art during processing of audio data sequences different in speed. Through the technical scheme provided by an embodiment of the invention, a same audio data sequence of an unknown type is divided at least two times to obtain a plurality of audio data subsequences different in length, and therefore, adaptability of the audio data sequence of an unknown type is enhanced, and the audio data sequence is applicable to different conditions like higher percentages of quicker audio data and slower audio data in a training sample. And therefore, the method is wider in application scope, and is fine in robustness for the audio data different in speed.

Description

technical field [0001] The invention relates to the technical field of information processing, in particular to an audio data classification method and device. Background technique [0002] The classification of audio data can be widely used in audio retrieval and abnormal event detection. For example, a specific example of performing audio retrieval on audio data is performing speech and music classification on audio data. In this example, if it can be determined in advance that the category of certain audio data to be retrieved is "music", it can be determined that the audio data should be searched in the audio data subcategory contained in the "music database". In particular, if it can be pre-determined that the type of audio data to be retrieved is audio data with a certain musical style, the retrieval scope can be further narrowed. [0003] Another example is to classify different audio data to be detected, such as screams, glass breaking, gunshots, and normal sounds ...

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 Patents(China)
IPC IPC(8): G06F17/30
CPCG06F16/683
Inventor 谢志明潘晖潘石柱张兴明傅利泉朱江明吴军吴坚
Owner ZHEJIANG DAHUA TECH CO LTD