Anda belum login :: 27 Nov 2024 08:02 WIB
Home
|
Logon
Hidden
»
Administration
»
Collection Detail
Detail
Gaining phonetic knowledge whilst improving syntetc speech quaIity.
Oleh:
Pols, L.C.W.
;
Bezoouen, R. van
Jenis:
Article from Journal - ilmiah internasional
Dalam koleksi:
Journal of Phonetics vol. 19 no. 1 (Jan. 1991)
,
page 139 - 146.
Ketersediaan
Perpustakaan PKBB
Nomor Panggil:
405/JOP/19
Non-tandon:
tidak ada
Tandon:
1
Lihat Detail Induk
Isi artikel
Speech synthesis-by-rule from unlimited text requires much phonetic knowledge. By trial-and-error, as well as by systematic diagnostic evaluation, system deficits can be identified and subsequently improved in the best possible way. Another approach is to tune the rules in accordance with global and/or specific facts derived from analyzing natural speech, accessible via a speech database. These two approaches may well be complementary, but they can also be in (partial) conflict. This is related to the fact that, frequently, rules do not just describe phonetic and linguistic facts but also serve to compensate for certain system artefacts or limitations, such as fixed-duration stressed diphones as building blocks, synchronous formant transitions, or an all-pole source-filter model. Ideally one would like to see more interaction between gaining phonetic knowledge and improving synthetic speech quality.
Opini Anda
Klik untuk menuliskan opini Anda tentang koleksi ini!
Kembali
Process time: 0.015625 second(s)