Confidence Measures for Keyword Spotting using Suport Vector Machines

Yassine Benayed; Dominique Fohr; Jean-Paul Haton; Gérard Chollet

Confidence Measures for Keyword Spotting using Suport Vector Machines

Yassine Benayed ¹ Dominique Fohr ¹ Jean-Paul Haton ¹ Gérard Chollet ²

1 PAROLE - Analysis, perception and recognition of speech

INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications

2 TSI - Laboratoire du traitement du signal et instrumentation

Abstract : Support Vector machines (SVM) is a new and very promising classification technique developed from the theory of Structural Risk Minimisation. In this paper, we propose an alternative out-of-vocabulary word detection method relying on confidence measures and support vector machines. Confidence measures are computed from phone level information provided by a Hidden Markov Model (HMM) based speech recognizer. We use three kinds of average techniques as arithmetic, geometric and harmonic averages to compute a confidence measure for each word. The acceptance/rejection decision of a word is based on the confidence feature vector which is processed by a SVM classifier. The performance of the proposed SVM classifier is compared with methods based on the averaging of confidence measures.

Mots-clés : détection de mot clé classifier svm keyword dectetion confidence measure classifieur svm mesure de confiance

Document type :

Conference papers

Domain :

Computer Science [cs] / Other [cs.OH]

Complete list of metadatas

https://hal.inria.fr/inria-00099706
Contributor : Publications Loria <>
Submitted on : Tuesday, September 26, 2006 - 9:40:29 AM
Last modification on : Thursday, November 21, 2019 - 2:12:33 AM

Confidence Measures for Keyword Spotting using Suport Vector Machines

Identifiers

Collections

Citation

Export

Metrics

270

Contact

Confidence Measures for Keyword Spotting using Suport Vector Machines

Identifiers

Collections

Citation

Export

Share

Metrics

270

Contact