Presentation | 2009-01-30 Statistical Voice Quality Transformation and Control Methods for Arbitrary Speakers Tomoki TODA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Voice conversion (VC) is a technique to modify nonlinguistic information such as voice characteristics while keeping linguistic information unchanged. In the traditional VC framework, we need to train a conversion model for a specific input and output speaker-pair using a parallel data set consisting of dozens of utterance-pairs of those two speakers. To make VC technologies more practical, we study a novel VC framework effectively using speech samples of many different speakers for flexibly developing the conversion models for arbitrary speakers. We have hitherto proposed many-to-one VC allowing the conversion from a specific speaker's voice into an arbitrary speaker's voice and one-to-many VC vice versa. This paper describes our attempts to develop these VC technologies. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | voice conversion / voice quality control / one-to-many conversion / many-to-one conversion / eigenvoices |
Paper # | SP2008-138 |
Date of Issue |
Conference Information | |
Committee | SP |
---|---|
Conference Date | 2009/1/22(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 | Speech (SP) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Statistical Voice Quality Transformation and Control Methods for Arbitrary Speakers |
Sub Title (in English) | |
Keyword(1) | voice conversion |
Keyword(2) | voice quality control |
Keyword(3) | one-to-many conversion |
Keyword(4) | many-to-one conversion |
Keyword(5) | eigenvoices |
1st Author's Name | Tomoki TODA |
1st Author's Affiliation | Graduate School of Information Science, Nara Institute of Science and Technology() |
Date | 2009-01-30 |
Paper # | SP2008-138 |
Volume (vol) | vol.108 |
Number (no) | 422 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |