Music retrieval method and system based on music fingerprint characteristic

Abstract

The invention discloses a music retrieval method and a music retrieval system based on a music fingerprint characteristic. The method comprises the steps of constructing a music fingerprint index, inputting an audio signal, performing preprocessing to obtain an audio frequency spectrogram, combining a first anchor point, a second anchor block and a third anchor block which are selected from the spectrogram to obtain a fingerprint, interpolating the fingerprint into a harsh table, and recording a song ID in which the fingerprint appears and the position of the fingerprint in a song; performing music retrieval: extracting a fingerprint sequence of a music fragment to be retrieved, searching the harsh table to obtain the song ID corresponding to each fingerprint and a song position chain table, constructing a matching table in which the song IDs correspond to the fingerprints, taking the most repeated positions in the song as a hit number in the matching table, and taking the song with the largest hit number as a retrieval result. According to the music retrieval method and the music retrieval system, quick retrieval is realized when a music library reaches the million level; the problem of channel drifting is solved; the noise resistance and the robustness of the fingerprints are improved; limitation to the shortest recording time is avoided; the fingerprint characteristic which is more compact and occupies a smaller memory space can be obtained.

Description

technical field [0001] The present invention relates to the technical field of audio signal processing and multimedia information retrieval system, and more specifically, to a music retrieval method and system based on music fingerprint features. Background technique [0002] With the advent of the Internet age, big data has increasingly become a hot topic. In the field of music retrieval, facing millions of music data, how to achieve fast retrieval has become a very valuable problem. In practical applications, the recording environment may be very noisy, and how to select noise-resistant and robust fingerprints has become a difficult problem. [0003] The existing music fingerprinting algorithms on the market include the music retrieval algorithm of Philips, see Haitsma J, Kalker T.A Highly Robust Audio Fingerprinting System[C]ISMIR.2002. It does this by taking a spectrogram of the music and extracting a feature for each frame. The spectrogram is divided into several fre...

AI Technical Summary

Problems solved by technology

The disadvantages of the Philips algorithm include: 1. The accuracy of feature extraction in the silent part is low

2. Due to serious channel distortion in the high and low frequency areas of recording equipment such as mobile phones, the accuracy rate is low

3. The music signal is actually sparse, and the Philips algorithm extracts features for each frame, resulting in redundancy of fingerprint features and slowing down retrieval

Shazam's fingerprint feature redundancy is lower than that of the Philips algorithm, but it has the following disadvantages: 1. When the music library reaches a million levels, due to the upper limit of the space limit of the buckets composed of point pairs, the search speed slows down

2. When the recording source is jittering or the recording channel is unstable, the anchor points and important points will drift, resulting in a decrease in recognition accuracy

3. There is a minimum recording time limit, such as at least 3 seconds of recording to be recognized

Images