Multi-speaker speech synthesis method based on probability generation and non-autoregression model

Abstract

The invention provides a multi-speaker speech synthesis method based on probability generation and a non-autoregression model. The method comprises the steps that S1, a speaker personalized encoder and a probability generation encoder receive a target Mel spectrum and extract a speaker personalized vector and a probability generation vector respectively; s2, encoding the spliced and fused input vectors based on an encoder of a deep network to obtain phoneme-level deep features; s3, a phoneme duration predictor receives the spliced and fused fusion features and predicts a phoneme duration sequence; s4, receiving the phoneme time length sequence and expanding the fusion feature by the length regulation network to obtain a frame-level feature; s5, a decoder based on the deep network receives the frame-level features and maps the frame-level features into a predicted Mel spectrum, and a post-processing network supplements residual information of the predicted Mel spectrum; and S6, the vocoder maps the predicted Mel spectrum supplemented with the residual information into a sound waveform to obtain synthetic speech. The generalization of a multi-speaker speech synthesis system and the similarity of synthesized speech can be improved.

Description

technical field [0001] The invention relates to the technical field of speech synthesis, in particular to a multi-speaker speech synthesis method based on probability generation and non-autoregressive models. Background technique [0002] Text To Speech (TTS) refers to a technology that can convert any text into audio. In recent years, the end-to-end single-speaker speech synthesis model based on deep learning has been able to synthesize clear and natural speech. With the speech synthesis With the further development of technology, its application scenarios are gradually increasing, and there is also a certain demand for multi-speaker speech synthesis technology, such as: rapid customization of sound library, audio novels, etc. [0003] The traditional multi-speaker speech synthesis (multi-speaker TTS) system uses a one-hot vector to represent the identity of the speaker, and synthesizes the speech of a specific speaker by changing the one-hot vector, but the one-hot vector ...

AI Technical Summary

Problems solved by technology

[0007] In order to solve the technical problems of insufficient generalization and low similarity of the multi-speaker speech synthesis system for speakers outside the data set in the prior art, the present invention proposes a multi-speaker speech synthesis system based on probability generation and non-autoregressive models. The speaker speech synthesis method is used to solve the above-mentioned technical problems to realize

Images