Speech Synthesis Method and System

Abstract

The present invention provides a speech synthesis method, which includes: acquiring an interaction instruction, and analyzing the interaction instruction, and responding to the speech playback instruction when the interaction instruction contains a speech playback instruction; invoking the acoustic The model is to input the text data to be synthesized or the phonetic symbol data to be synthesized into the acoustic model in real time to obtain the acoustic features to be synthesized; input the acoustic features to be synthesized into the acoustic synthesizer, obtain the synthesized voice data through the acoustic synthesizer, and output the voice data. The invention provides an end-to-end acoustic model and a training method, which can fully utilize the advantages of parallel computing. Using a fully parallel attention mechanism and a convolutional structure, the time to generate acoustic features is greatly shortened compared with the prior art, and at the same time, the quality of the generated acoustic features can be guaranteed. Therefore, the present invention can shorten the synthesis time while ensuring the quality of synthesized voice, and improve the efficiency of voice synthesis.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a speech synthesis method and system. Background technique [0002] Speech synthesis, also known as text-to-speech (Text to Speech) technology, can convert any text information into a standard and smooth voice in real time. It involves many disciplines and technologies such as acoustics, linguistics, digital signal processing, and computer science. The main problem to be solved is how to convert text information into audible sound information. [0003] Currently, the data used for speech synthesis training generally consists of plain text (or phonetic symbols) and corresponding audio. There is no boundary information of phonetic symbols in text (or phonetic symbols), which makes it very difficult to determine the sequence correspondence between acoustic features and text (phonetic symbols). Moreover, the current acoustic training model can only generate acoustic features ...

AI Technical Summary

Problems solved by technology

There is no boundary information of phonetic symbols in the text (or phonetic symbols), which makes it very difficult to determine the sequence correspondence between the acoustic features and the text (phonetic symbols)

Moreover, the current acoustic training model can only generate acoustic features one time period at a time, and cannot make full use of the advantages of parallel computing, making its time delay very long

Images