Lightweight multi-speaker voice synthesis system and electronic equipment

Abstract

The invention discloses a lightweight multi-speaker voice synthesis system and electronic equipment. The system comprises a text feature extraction and normalization module, a speaker feature extraction module, a feature fusion module and a voice generation module. The text feature extraction and normalization module is used for carrying out encoding and feature extraction on to-be-processed textinformation by adopting a lightweight encoder, carrying out duration prediction on each word or phoneme corresponding to text deep features output by the lightweight encoder by adopting a lightweightduration prediction network, and carrying out length normalization processing to obtain regular text features with length equal to that of a target Mel spectrum. The speaker feature extraction moduleis used for generating features capable of representing a target speaker. The feature fusion module is used for fusing the features of the target speaker with the regular text features. The voice generation module is used for carrying out deep feature extraction, dimension mapping and residual error integration on the fused features and generating voice. The system supports multi-speaker voice synthesis and is high in synthesis speed.

Description

technical field [0001] The disclosure belongs to the technical field of speech synthesis, and relates to a lightweight multi-speaker speech synthesis system and electronic equipment. Background technique [0002] In recent years, neural network-based end-to-end speech synthesis systems have surpassed traditional statistical parametric speech synthesis systems in terms of system architecture and generated speech quality. End-to-end speech synthesis systems, such as Tacotron2 system and Transformer text-to-speech system (Transformer TTS system for short), directly use the neural network to convert text into corresponding speech, no longer need a lot of complicated text front-end processing work, various Extraction of linguistic features and complex domain expert knowledge. [0003] However, most of the current mainstream end-to-end speech synthesis systems use the attention mechanism to implicitly learn the text-to-speech alignment relationship, which brings a huge amount of ...

AI Technical Summary

Problems solved by technology

[0005] In addition, most of the existing text-to-speech systems can only realize a single speaker's single-style speech synthesis, and a few speech synthesis systems that can realize multi-speaker synthesis have the disadvantages of slow synthesis speed, large amount of calculation and memory consumption. question

Images