Method for synthesizing pronunciation based on rhythm model and parameter selecting voice

Abstract

The invention provides a voice synthesizing method based on rhythm model and parameter-based sound selection, making acoustics rhythm parameter planning to obtain the target values of acoustics parameters expected by each syllable; then making maximum matching, selecting those with the smallest difference as really used samples; after the maximum matching, making single character matching treatment on the unmatched segments; calculating a synthesizing cost of all segment paths through all syllable candidate samples, where the synthesizing cost is determined by the difference between the acoustics parameters of each candidate sample and their planned values and the difference synthesis between the candidate samples of two adjacent syllables in the paths; obtaining a path with the lowest synthesizing cost by dynamical planning algorithm; as all syllable samples are selected, obtaining the data in a voice base and making waveform splicing and obtaining the final synthesis result.

Description

technical field [0001] The invention relates to the technical field of speech synthesis, in particular to a speech synthesis method. Background technique [0002] At present, the development direction of Chinese speech synthesis is based on the waveform splicing technology of large-scale real recording sound library. The so-called large-scale real recording sound library refers to the recording sound library that records a large number of natural speech, and its scope basically covers various pronunciation situations in most contexts. For different contexts, the system will select the most matching Original speech fragments to be spliced. Due to the large size of the sound library, in almost all cases, the most suitable original natural speech can be found without using other techniques for adjustment, thus ensuring the consistency of the final synthesized speech and the original speech. In addition, the fragments selected here go beyond the level of syllables, and can be ...

AI Technical Summary

Problems solved by technology

For example, if two consecutive syllables in a sentence are selected from the two sentences respectively, although the sound selection is carried out according to the position, the actual acoustic parameters may not be considered, which may cause the two consecutive syllables not to conform to the actual voice. law of change

This leads to a pitch jump in the sense of hearing, or a mismatch in the length of the sound, which reduces the naturalness of the voice

Images