Audio single tone separation method, audio single tone separation device, computer equipment and storage medium

Abstract

The invention discloses an audio single tone separation method, an audio single tone separation device, computer equipment and a storage medium. The invention is applied to the technical field of audio processing, and is used for solving the problem that single tone separation cannot be realized in the prior art. The method provided by the invention comprises the following steps: acquiring a target audio to be subjected to audio separation; determining each tone type required to be separated for the target audio; selecting one LSTM neural network corresponding to each tone type from each pre-trained LSTM neural network, wherein the LSTM neural networks serve as target LSTM neural networks, each LSTM neural network is obtained by pre-training audio samples corresponding to different timbretype combinations, and each timbre type combination is composed of more than two timbre types; inputting the target audio as input into a target LSTM neural network to obtain each output target spectrogram; and performing time domain transformation on each target spectrogram to obtain a target single-tone audio corresponding to each target spectrogram, and taking the target single-tone audio as anaudio separation result of the target audio.

Description

technical field

[0001] The invention relates to the technical field of audio processing, in particular to an audio monochrome separation method, device, computer equipment and storage medium. Background technique

[0002] In the development of music library, music content analysis is very important. Generally, the audio collected naturally is often mixed with multiple musical instruments and human voices. If the single-tone color separation of various musical instruments and human voices in the audio can be achieved, it can provide strong material support for the content analysis of single musical instruments and human voices. , and promote the rapid development of music pitch recognition technology and automatic transcription technology. It can be seen that it is of great significance and value to achieve monochrome separation for audio.

[0003] Therefore, finding a method capable of realizing audio monochrome separation has always been an urgent problem to be solved by ...

Images