Confidence estimation, OOV detection and language ID using phone-to-word transduction and phone-level alignments
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F08%3APU76780" target="_blank" >RIV/00216305:26230/08:PU76780 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Confidence estimation, OOV detection and language ID using phone-to-word transduction and phone-level alignments
Original language description
Automatic Speech Recognition (ASR) systems continue to make errors during search when handling various phenomena including noise, pronunciation variation, and out of vocabulary (OOV) words. Predicting the probability that a word is incorrect can preventthe error from propagating and perhaps allow the system to recover. This paper addresses the problem of detecting errors and OOVs for read Wall Street Journal speech when the word error rate (WER) is very low. It augments a traditional confidence estimate by introducing two novel methods: phone-level comparison using Multi-String Alignment (MSA) and word-level comparison using phone-to-word transduction. We show that features from phone and word string comparisons can be added to a standard maximum entropy framework thereby substantially improving performance in detecting both errors and OOVs. Additionally we show an extension to detecting English and accented English for the Language Identification (LID) task.
Czech name
Odhad spolehlivosti, detekce OOV a identifikace jazyka pomocí transducerů převádějících fonémy na slova a fonetických zarovnání
Czech description
Článek je o odhadu spolehlivosti, detekci OOV a identifikaci jazyka pomocí transducerů převádějících fonémy na slova a fonetických zarovnání<br>
Classification
Type
D - Article in proceedings
CEP classification
JC - Computer hardware and software
OECD FORD branch
—
Result continuities
Project
—
Continuities
Z - Vyzkumny zamer (s odkazem do CEZ)
Others
Publication year
2008
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. 2008 IEEE International Conference on Acoustics, Speech, and Signal Processing
ISBN
1-4244-1484-9
ISSN
—
e-ISSN
—
Number of pages
4
Pages from-to
—
Publisher name
IEEE Signal Processing Society
Place of publication
Las Vegas
Event location
Las Vegas
Event date
Mar 30, 2008
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—