Confidence measure improvement using useful predictor features and support vector machines

In traditional keyword spotting (KWS) systems, confidence measure (CM) of each keyword is computed from normalized acoustic likelihoods. In addition to likelihood based scores, some keyword dependent features named predictor features such as duration and prosodic features could be defined to improve the performance of CM. In this paper a discriminative and probabilistic computation of CM based upon some useful predictor features and support vector machines (SVM) is presented for Persian conversational telephone speech KWS. Our experimental results show that higher performance will be achieved by appending utilized predictor features. The proposed CM with linear kernel function of SVM is obtained an improvement about 8.6% in Figure-of-Merit (FOM) of KWS system.

[1]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[2]  Gérard Chollet,et al.  Confidence measures for keyword spotting using support vector machines , 2003, 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)..

[3]  Rong Zhang,et al.  Word level confidence annotation using combinations of features , 2001, INTERSPEECH.

[4]  Dong Wang,et al.  Augmented set of features for confidence estimation in spoken term detection , 2010, INTERSPEECH.

[5]  Ahmad Akbari,et al.  Performance evaluation for an HMM-based keyword spotter and a large-margin based one in noisy environments , 2011, WCIT.

[6]  Gérard Chollet,et al.  Keyword Spotting Using Support Vector Machines , 2002, TSD.

[7]  Richard Rose,et al.  A hidden Markov model based keyword recognition system , 1990, International Conference on Acoustics, Speech, and Signal Processing.

[8]  Mahmood Bijankhan,et al.  Tfarsdat - the telephone farsi speech database , 2003, INTERSPEECH.

[9]  Jan Cernocký,et al.  Acoustic keyword spotter - optimization from end-user perspective , 2010, 2010 IEEE Spoken Language Technology Workshop.

[10]  Kate Knill,et al.  Fast implementation methods for Viterbi-based word-spotting , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[11]  Hui Jiang,et al.  Confidence measures for speech recognition: A survey , 2005, Speech Commun..

[12]  Jia Liu,et al.  A New Framework For Large Vocabulary Keyword Spotting Using Two-Pass Confidence Measure , 2006, The Proceedings of the Multiconference on "Computational Engineering in Systems Applications".