Speech synthesis method and related equipment

Abstract

The present application provides a speech synthesis method and related equipment. The method includes the following steps that: the identity of a user is determined according to the current input speech of the user; an acoustic model is obtained from an acoustic model library according to the current input speech; basic speech synthesis information is determined according to the identity of the user, wherein the basic speech synthesis information characterizes variable quantities in the preset sound speed, preset volume, and preset pitch of the acoustic model; a reply text is determined; enhanced speech synthesis information is determined according to the reply text and context information, wherein the enhanced speech synthesis information characterizes variable quantities in the preset timbre, tone and preset rhythm of the acoustic model; and speech synthesis is performed on the reply text through the acoustic model according to the basic speech synthesis information and the enhancedspeech synthesis information, so that reply speech for the user can be obtained. With the speech synthesis method and related apparatus provided by the embodiments of the invention adopted, a device can provide a personalized speech synthesis effect to the user during a man-machine interaction process, and therefore, the speech interaction experience of the user can be improved.

Description

technical field [0001] The invention relates to the field of speech processing, in particular to a speech synthesis method and related equipment. Background technique [0002] In recent years, human-computer dialogue has widely entered people's daily lives. Common scenarios include intelligent customer service robots, smart speakers, and chat robots. The core of human-machine dialogue is that the machine can automatically understand and analyze the voice input by the user under the framework of the built system and according to the data of prior training or learning, and give a meaningful voice reply. When designing a text-to-speech system for Chinese text, if only the input text is matched to the pronunciation database one by one, and the pronunciation of all characters is concatenated to form a voice output, then such a voice will be very mechanical and blunt without intonation fluctuations. The listening experience is poor. The TTS (text–to-speed) engine developed in re...

AI Technical Summary

Problems solved by technology

[0003] Today, the existing technology is only limited to making the machine "speak like a human" in the process of human-computer interaction, without considering the diverse needs of users for human-computer interaction

Images