DocumentCode
2229169
Title
Word graph rescoring using confidence measures
Author
Fetter, Pablo ; Dandurand, Frédéric ; Regel-Brietzmann, Peter
Author_Institution
Res. & Technol., Daimler-Benz AG, Ulm, Germany
Volume
1
fYear
1996
fDate
3-6 Oct 1996
Firstpage
10
Abstract
This paper presents a novel approach to using confidence scores for word graph rescoring. For each word in the system´s vocabulary we computed the probability that the observation is correct given its acoustic score. Afterwards, we used these probabilities for rescoring word graphs outputted by the recognizer. We present some implementation detail as well as accuracy improvements obtained using this method
Keywords
graph theory; hidden Markov models; natural language interfaces; probability; speech recognition; vocabulary; acoustic score; confidence measures; hidden Markov model; probability; speech recognition; vocabulary; word graph rescoring; Bayesian methods; Distributed computing; Hidden Markov models; Military computing; Pattern recognition; Speech recognition; Tail; Technical Activities Guide -TAG; Testing; Vocabulary;
fLanguage
English
Publisher
ieee
Conference_Titel
Spoken Language, 1996. ICSLP 96. Proceedings., Fourth International Conference on
Conference_Location
Philadelphia, PA
Print_ISBN
0-7803-3555-4
Type
conf
DOI
10.1109/ICSLP.1996.606917
Filename
606917
Link To Document