Voice isolation device, voice synthesis device, and voice quality conversion device

Abstract

A voice separation device includes: a PARCOR calculation unit (102) which extracts vocal information from an input voice signal; a filter smoothing unit (103) which smoothes the vocal information extracted by the PARCOR calculation unit (102) within a first time band; a reverse filter unit (104) which calculates a filter coefficient of a filter having a reverse frequency amplitude response characteristic of vocal information which has been smoothed by the filter smoothing unit (103); and a sound source modeling unit (105) which cuts out a waveform contained in a second time band shorter than the first time band from the input voice signal which has been filtered by the reverse filter unit (104) and calculates sound source information from the waveform.

Description

Technical field [0001] The present invention relates to a sound separation device, a sound synthesis device, and a sound quality conversion device that separate an input sound signal into sound source information and channel information. Background technique [0002] In recent years, with the development of sound synthesis technology, it has become possible to produce very high-quality synthetic sounds. [0003] However, the conventional use of synthesized speech is centered on the unified use of reading news articles in the tone of the announcer. [0004] On the other hand, mobile phone services, etc., provide services that use celebrity voice messages instead of incoming call voices, etc., with characteristic voices (synthetic sounds with high personal reproducibility, high school girl style or Kansai dialect style, etc. The characteristic rhythm and sound quality of the synthesizing sound) began to circulate as one content. It is conceivable that in order to increase the fun i...

AI Technical Summary

Problems solved by technology

As a result, there is a problem in that fast motion that is not present in the original sound channel is used as channel information, and fast motion that is present in the original sound source is removed from the sound source information.

However, there is a problem that the sound quality cannot be changed by controlling the channel information and the sound source information independently, like only the channel information or only the sound source information.

Therefore, there is a problem that the sound quality cannot be changed arbitrarily in this state