TRAP based features for LVCSR of meeting data
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F04%3APU49150" target="_blank" >RIV/00216305:26230/04:PU49150 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
TRAP based features for LVCSR of meeting data
Original language description
This paper describes using temporal patterns (TRAPs) feature extraction in large vocabulary continuous speech recognition (LVCSR) of meeting data. Frequency differentiation and local operators are applied to critical-band speech spectrum. Testsare performed with HMM recognizer on ICSI meetings database. We show that TRAP features in combination with standard ones lead to improvement of word-error rate (WER).
Czech name
Priznaky na bazi TRAP pro LVCSR meetingovych dat
Czech description
Tento clanek popisuje pouziti TRAP priznaku v LVCSR. Differenciace ve spektru a localni operatory jsou applikovany na recove spektrum. Testy jsou provadeny s HMM rozpoznavacem na ICSI meetings databazi. Dokazujeme ze TRAP priznaky v kombinaci se standartnimy priznaky vedou ke zlepseni uspesnosti rozpoznavace.<br>
Classification
Type
D - Article in proceedings
CEP classification
JC - Computer hardware and software
OECD FORD branch
—
Result continuities
Project
<a href="/en/project/GP102%2F02%2FD108" target="_blank" >GP102/02/D108: Data-driven and anthropic speech coding and recognition</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2004
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Proc. 8th International Conference on Spoken Language Processing
ISBN
—
ISSN
—
e-ISSN
—
Number of pages
4
Pages from-to
437-440
Publisher name
Sunjin Printing Co,
Place of publication
Jeju Island
Event location
Jeju Island
Event date
Oct 4, 2004
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—