Audio frequency feature library establishing method and device

A technology of audio features and establishment methods, applied in speech analysis, speech recognition, instruments, etc., can solve the problems of reducing the performance of music recognition system and increasing the difficulty of matching, so as to alleviate the mismatch phenomenon, offset channel distortion, and improve accuracy. rate effect

Active Publication Date: 2013-09-04
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF6 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in practical applications, the signal recorded by the user will have obvious interference, including the system noise introduced by the playback device, recording device, etc., and the noise of the surrounding environment of the recording, and the signal used for training is generally pure. Music files (such as MP3, APE and other audio formats), which makes it more difficult to match in real application environments, thereby reducing the performance of the music recognition system

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 frequency feature library establishing method and device
  • Audio frequency feature library establishing method and device
  • Audio frequency feature library establishing method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0057]In order to realize the search of audio content, it is necessary to use a large-scale music library for training and establish an audio feature library, such as the existing CMI feature training technology, which mainly includes four steps:

[0058] 1) Audio feature extraction. For each audio file in the music library, features such as music melody and rhythm are extracted frame by frame;

[0059] 2) Audio segmentation, looking for the mutation points of the audio signal, using these mutation points to divide the training data into several audio segments (Segments), and extracting a feature vector from each Segment;

[0060] 3) Feature clustering, through a specific clustering algorithm, the features of each segment are clustered, and the most representative K-type features of a piece of music are extracted while reducing the number of features, K is the number of clusters;

[0061] 4) Index establishment, to establish an index table, such as a hash table, for the featur...

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 an audio frequency feature library establishing method and device. The audio frequency feature library establishing method comprises the steps of estimating the noise feature of a recording and playing system; according to the estimated noise feature of the recording and playing system, carrying out imnoise processing on audio frequencies in an audio frequency library; extracting features of the audio frequencies which are subjected to imnoise processing; and an audio frequency feature library is established through the extracted features. According to the technical scheme of the audio frequency feature library provided by the embodiment of the invention, comparing with the training method of a traditional CMI (Contend-based Music Identification) system, the imnoise processing is added in the training phase, the audio frequency feature library which is obtained through the imnoise processing is used to relieve the mismatch phenomenon of training signals and test signals, and the accuracy of an audio frequency identification system can be effectively improved.

Description

technical field [0001] The invention relates to the technical field of audio processing, in particular to a method and device for establishing an audio feature library. Background technique [0002] With the development of the Internet, the objects that users search on the Internet are not limited to text content, and pictures, audio, video, etc. have all become objects supported by search engines. For example, CMI (Contend-based Music Identification, content-based music identification) is a popular application form in the Internet at present. In terms of form, this application is similar to traditional text search. When users hear a piece of music that they are interested in but do not know the title of the song, they can record a few seconds of music and submit the piece as a search request to the corresponding Audio search system, the system finds various information of the music through the background search technology and feeds back to the user. [0003] In order to r...

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): G10L15/06
Inventor 宋辉
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products