Text-To-Speech synthesis using HMM-based triphones

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F99%3A00044605" target="_blank" >RIV/49777513:23520/99:00044605 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Text-To-Speech synthesis using HMM-based triphones
Original language description
Automatic construction of a Czech speech segment database (SSD) from a single speaker natural speech corpus using multiple Hidden Markov Models (HMMs) and an approach to speech synthesis based on this inventory are presented in this paper. HMMs are usedto model triphones. Binary decision trees are employed to automatically cluster the states of triphones. Resulting models are used to segment the speech corpus and to create a SSD. The SSD constructed in this way is assumed to enable more precise contextmodeling than was previously possible. Several speech synthesis techniques are discussed to design a concatenation-based synthesizer. We will focus on an MFCC-based pitch-synchronous residually excited approach.
Czech name
—
Czech description
—

Project
<a href="/en/project/VS97159" target="_blank" >VS97159: Center for research in the field of cybernetic systems</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>Z - Vyzkumny zamer (s odkazem do CEZ)

Publication year
1999
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Text-To-Speech synthesis using HMM-based triphones
ISBN
—
ISSN
—
e-ISSN
—
Number of pages
1
Pages from-to
—
Publisher name
Digital Media Services in conjuction with DSP World
Place of publication
Orlando
Event location
—
Event date
—
Type of event by nationality
—
UT code for WoS article
—

Similar results(10)