DocumentCode
1937143
Title
Compression of acoustic inventories using asynchronous interpolation
Author
Kain, Alexander B. ; Van Santen, Jan P H
Author_Institution
Center for Spoken Language Understanding, OGI Sch. of Sci. & Eng. at OHSU, Beaverton, OR, USA
fYear
2002
fDate
11-13 Sept. 2002
Firstpage
83
Lastpage
86
Abstract
A compression method is proposed that takes advantage of a powerful property of acoustic unit inventories: In the appropriate acoustic space, units that share a (context-dependent or -independent) phoneme label must be close to a vector phoneme template associated with the phoneme. The method approximates units by interpolation between templates. The interpolation operation involves two asynchronous weight functions operating on the template. One is associated with spectral peak locations, the second with spectral balance. This enables approximating transitions such as [i:]→[v], in which formant movement precedes frication onset. The algorithm guarantees smooth concatenation points.
Keywords
acoustic signal processing; data compression; spectra; spectral analysis; speech coding; speech synthesis; acoustic inventories; approximating transitions; asynchronous interpolation; asynchronous weight functions; compression method; phoneme label; smooth concatenation points; spectral balance; vector phoneme template; Acoustic applications; Acoustic devices; Acoustical engineering; Artificial intelligence; Compression algorithms; Interpolation; Loudspeakers; Natural languages; Speech coding; Speech synthesis;
fLanguage
English
Publisher
ieee
Conference_Titel
Speech Synthesis, 2002. Proceedings of 2002 IEEE Workshop on
Print_ISBN
0-7803-7395-2
Type
conf
DOI
10.1109/WSS.2002.1224378
Filename
1224378
Link To Document