Method and equipment for voice synthesis and method for training acoustic model used in voice synthesis

Abstract

The invention relates to a method and equipment for voice synthesis and a method for training an acoustic model used in voice synthesis. The method for voice synthesis includes the steps as follows: confirming that data generated by text analysis is fuzzy polyphone data; and performing fuzzy polyphone prediction for the fuzzy polyphone data, so as to output a plurality of candidate pronunciations and the probability thereof; generating the fuzzy context characteristic tagging based on the candidate pronunciations and the probability thereof; based on the acoustical model provided with a fuzzy decision tree, confirming model parameters direct at the fuzzy context characteristic tagging; generating voice parameters based on the model parameters; and synthesizing voice through the voice parameters. As per the method and equipment provided by the embodiment of the invention, the fuzzy treatment can be performed for polyphone words difficult for prediction in a Chinese text, so as to improve the synthesis quality of Chinese polyphones.

Description

technical field [0001] The present invention relates to speech synthesis, more specifically, to the synthesis of Chinese polyphonic characters. Background technique [0002] The speech generated by humans through certain machine equipment is called speech synthesis. Speech synthesis is an important part of human-machine speech communication. The use of speech synthesis technology can make the machine speak like a human, so that some information expressed or stored in other ways can be converted into speech, so that people can easily obtain this information through hearing. [0003] At present, a lot of research and application is carried out on the text-to-speech TTS system. In this system, the text to be synthesized is usually input, and the text analyzer included in the system processes it, and outputs pronunciation descriptors, which include phonetic symbols and phonetic symbols at the segment level. Prosodic notation at the suprasegmental level. The text analyzer firs...

AI Technical Summary

Problems solved by technology

If the speech synthesis system gives the wrong pronunciation, it will cause ambiguity in the listener's understanding and give the listener a very bad feeling

Thus, for the speech synthesis system applied in life, work and scientific research (such as car navigation, automatic audio service, broadcasting, robot simulation, etc.), it will cause bad user experience due to obviously wrong pronunciation of polyphonic characters, and even use the inconvenience

Images