Details
| Originalsprache | Englisch |
|---|---|
| Seiten (von - bis) | 1084-1095 |
| Seitenumfang | 12 |
| Fachzeitschrift | Acta Acustica united with Acustica |
| Jahrgang | 90 |
| Ausgabenummer | 6 |
| Publikationsstatus | Veröffentlicht - Nov. 2004 |
Abstract
Progress mae with the AT&T sample-based visual text-to-speech (VTTS) system is discussed. The VTTS system from AT&T incorporates unit selection synthesis and a moderate size recorded database of modified and concatenated video segments. It is suggested that several steps such as highly accurate image analysis tools for creating video clip databases, fast research techniques and rendering of composite face images on a graphic screen are very important to assure a high quality sample based VTTS system. It was found that accuracy and timeliness of lip closures and protrusions, turning points and overall smoothness are very critical for the system.
ASJC Scopus Sachgebiete
- Geisteswissenschaftliche Fächer (insg.)
- Musik
- Physik und Astronomie (insg.)
- Akustik und Ultraschall
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
in: Acta Acustica united with Acustica, Jahrgang 90, Nr. 6, 11.2004, S. 1084-1095.
Publikation: Beitrag in Fachzeitschrift › Artikel › Forschung › Peer-Review
}
TY - JOUR
T1 - From audio-only to audio and video text-to-speech
AU - Cosatto, Eric
AU - Graf, Hans Peter
AU - Ostermann, Jörn
AU - Schroeter, Juergen
PY - 2004/11
Y1 - 2004/11
N2 - Progress mae with the AT&T sample-based visual text-to-speech (VTTS) system is discussed. The VTTS system from AT&T incorporates unit selection synthesis and a moderate size recorded database of modified and concatenated video segments. It is suggested that several steps such as highly accurate image analysis tools for creating video clip databases, fast research techniques and rendering of composite face images on a graphic screen are very important to assure a high quality sample based VTTS system. It was found that accuracy and timeliness of lip closures and protrusions, turning points and overall smoothness are very critical for the system.
AB - Progress mae with the AT&T sample-based visual text-to-speech (VTTS) system is discussed. The VTTS system from AT&T incorporates unit selection synthesis and a moderate size recorded database of modified and concatenated video segments. It is suggested that several steps such as highly accurate image analysis tools for creating video clip databases, fast research techniques and rendering of composite face images on a graphic screen are very important to assure a high quality sample based VTTS system. It was found that accuracy and timeliness of lip closures and protrusions, turning points and overall smoothness are very critical for the system.
UR - http://www.scopus.com/inward/record.url?scp=11244348117&partnerID=8YFLogxK
M3 - Article
AN - SCOPUS:11244348117
VL - 90
SP - 1084
EP - 1095
JO - Acta Acustica united with Acustica
JF - Acta Acustica united with Acustica
SN - 1610-1928
IS - 6
ER -