| Presentation | 2007/12/13 F0 Gradient Model for Acoustic Quality and F0 Consistency of Concatenative TTS Ryuki TACHIBANA, Tohru NAGANO, Masafumi NISHIMURA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | A problem with concatenative text-to-speech synthesis is that it sometimes fails to preserve the appropriate consistency in the F0 contours at the concatenation points of the speech segments. Since Japanese is a pitch accent language, listeners perceive inconsistency in F0 contours as strange accents or wrong accent nuclei. Such problems occur more frequently when the database size is limited or when synthesizing voices for texts in new application domains. In this paper, we propose an F0 gradient model and F0 adjustment to select consistent speech segments and to restore the consistency by adjusting the F0 values only where necessary. This makes it possible to generate synthetic voices with correct pitch accents while taking advantages of the acoustic quality of the recorded speech segments even in new application domains. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Concatenative Text-to-Speech Synthesis / Prosody / Fundamental Frequency / Pitch Accent |
| Paper # | NLC2007-76,SP2007-139 |
| Date of Issue |
| Conference Information | |
| Committee | NLC |
|---|---|
| Conference Date | 2007/12/13(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Natural Language Understanding and Models of Communication (NLC) |
|---|---|
| Language | ENG |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | F0 Gradient Model for Acoustic Quality and F0 Consistency of Concatenative TTS |
| Sub Title (in English) | |
| Keyword(1) | Concatenative Text-to-Speech Synthesis |
| Keyword(2) | Prosody |
| Keyword(3) | Fundamental Frequency |
| Keyword(4) | Pitch Accent |
| 1st Author's Name | Ryuki TACHIBANA |
| 1st Author's Affiliation | Tokyo Research Lab., IBM Japan() |
| 2nd Author's Name | Tohru NAGANO |
| 2nd Author's Affiliation | Tokyo Research Lab., IBM Japan |
| 3rd Author's Name | Masafumi NISHIMURA |
| 3rd Author's Affiliation | Tokyo Research Lab., IBM Japan |
| Date | 2007/12/13 |
| Paper # | NLC2007-76,SP2007-139 |
| Volume (vol) | vol.107 |
| Number (no) | 405 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |