Audio file retrieving method and system as well as audio file type identification method and system

Abstract

The invention provides an audio file retrieving method and an audio file retrieving system as well as an audio file type identification method and an audio file type identification system. The audio file retrieving method comprises the following steps of: pre-determining the Mel frequency ceptral coefficient (MFCC) and / or the melody feature Pitch coefficient of each audio file in an audio file bank; extracting the MFCC and / or the melody feature Pitch coefficient of the audio file to be retrieved; matching the MFCC and / or the melody feature Pitch coefficient of the audio file to be retrieved with the MFCC and / or the melody feature Pitch coefficient of each audio file in the audio file bank; and retrieving the audio files from the audio file bank according to the matching results. When the methods and the systems are applied, the audio files can be retrieved according to the melody, or the audio file in a certain type can be retrieved, or the efficiency for identifying the types of the audio files is improved.

Description

Technical field [0001] The invention relates to the technical field of audio file retrieval, in particular to an audio file retrieval method and system, and an audio file type identification method and system. Background technique [0002] The core task of audio file retrieval is to retrieve audio files that meet user needs based on the audio file library. [0003] At present, the audio file retrieval system can only retrieve audio files based on the audio file name, singer name and other attributes of the audio file provided by the user. This method has the following disadvantages: [0004] First, when the user does not know the attributes and characteristics of the audio file, such as the name of the singer, and can only hum a tune, the existing audio retrieval system cannot retrieve the corresponding audio file according to the tune. [0005] Furthermore, when the user is not looking for a certain audio file, but for an audio file that can express a certain mood, the existing audio...

AI Technical Summary

Problems solved by technology

[0004] First of all, when the user does not know the attribute characteristics such as the name of the audio file and the name of the singer, but can only hum the tune, the existing audio retrieval system cannot retrieve the corresponding audio file according to the tune.

[0005] Furthermore, when the user does not want to find a certain audio file, but wants to find an audio file that can express a certain artistic conception, the existing audio retrieval system cannot retrieve the corresponding audio file.

[0006] In addition, if a manual method is used to identify the artistic conception type for all audio files, it will consume a lot of manpower and the efficiency is low

Images