Audio processing method, vocoder, device, equipment and storage medium

Abstract

The invention provides an audio processing method, a vocoder, a device, equipment and a storage medium. The method comprises the following steps: performing speech feature conversion on a to-be-processed text to obtain at least one acoustic feature frame; extracting condition features from each acoustic feature frame through a frame rate network; performing frequency band division and time domain downsampling on the current frame to obtain n subframes containing a preset number of sampling points; synchronously performing sampling value prediction on the current m adjacent sampling points corresponding to the n sub-frames in the ith round of prediction process through the sampling prediction network to obtain m * n sub-prediction values, and further obtaining n sub-prediction values corresponding to each sampling point in the preset number of sampling points; obtaining an audio prediction signal corresponding to the current frame according to the n sub-prediction values corresponding to each sampling point; and performing audio synthesis on each acoustic feature frame of the at least one acoustic feature frame to obtain a target audio. According to the invention, the audio processing speed and efficiency can be improved.

Description

Technical field [0001] The present application relates to audio and video processing techniques, and more particularly to an audio processing method, a vocoder, a device, a device, and a storage medium. Background technique [0002] With the rapid development of smart equipment (such as smartphone, smart speakers, etc.), speech interaction technology is a growing application as a natural interaction. As an important part of speech interaction technology, speech synthesis technology has also achieved great progress. Voice synthesis technology converts text into a corresponding audio content through a certain rule or model algorithm. Traditional speech synthesis is based primarily based on splicing methods or statistical parameters. With the continuous breakthrough in the field of speech recognition, deep learning is gradually introduced into the field of speech synthesis. Thanks to this, the neural network-based vocoder (Neural Vocoder) has made great progress. However, the curren...

AI Technical Summary

Problems solved by technology

However, the current vocoder usually needs to perform multiple cycles based on multiple sampling time points in the audio feature signal to complete speech prediction, and then complete speech synthesis, which leads to slower processing speed of audio synthesis and reduces the efficiency of audio processing

Images