| Presentation | 2004/3/18 Subjective Evaluation of Synthetic Utterance Animation in Acoustically Noisy Enviroments Akinobu MAEJIMA, Tatsuo YOTSUKURA, Shigeo MORISHIMA, Satoshi NAKAMURA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Authors have proposed a synthesis method of natural talking face. However, an evaluation method of talking animation quality is normally based on a subjective test. In this paper, a new evaluation test method of talking face including an objective evaluation is proposed. Quality of talking face will be evaluated by following three factors. Lip reading is possible? Visually natural? Synchronizing with speech? Lip reading quality is evaluated by the rate of correct answer of words when taling face and speech are presented to a subject under acoustically noisy environment and getting a listener's answer about uttered contents of speech. Secondary, visually naturalness of talking face and smoothness of lip movement are evaluated by five-level MOS rating. Finally, talking face and speech signal are presented asynchronously, then a subjective score depending on starting time difference between image and audio track is investigated. The quality of our synthetic talking face is evaluated by these methods and natural synchronization between a synthetic talking face and speech are verified. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Synthetic Talking Face / Acoustically Noisy Enviroments / Digit Discrimination Rate / Representation / Naturalization / Lip-Synchronization |
| Paper # | HCS2003-66 |
| Date of Issue |
| Conference Information | |
| Committee | HCS |
|---|---|
| Conference Date | 2004/3/18(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 | Human Communication Science (HCS) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Subjective Evaluation of Synthetic Utterance Animation in Acoustically Noisy Enviroments |
| Sub Title (in English) | |
| Keyword(1) | Synthetic Talking Face |
| Keyword(2) | Acoustically Noisy Enviroments |
| Keyword(3) | Digit Discrimination Rate |
| Keyword(4) | Representation |
| Keyword(5) | Naturalization |
| Keyword(6) | Lip-Synchronization |
| 1st Author's Name | Akinobu MAEJIMA |
| 1st Author's Affiliation | Faculty of Engineering, Seikei University:ATR Spoken Language Translation Research Laboratory() |
| 2nd Author's Name | Tatsuo YOTSUKURA |
| 2nd Author's Affiliation | ATR Spoken Language Translation Research Laboratory |
| 3rd Author's Name | Shigeo MORISHIMA |
| 3rd Author's Affiliation | Faculty of Engineering, Seikei University:ATR Spoken Language Translation Research Laboratory |
| 4th Author's Name | Satoshi NAKAMURA |
| 4th Author's Affiliation | ATR Spoken Language Translation Research Laboratory |
| Date | 2004/3/18 |
| Paper # | HCS2003-66 |
| Volume (vol) | vol.103 |
| Number (no) | 742 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |