
For FullText PDF, please login, if you are a member of IEICE,
or go to Pay Per View on menu list, if you are a nonmember of IEICE.

Utterance Verification Using StateLevel LogLikelihood Ratio with Frame and State Selection
SukBong KWON Hoirin KIM
Publication
IEICE TRANSACTIONS on Information and Systems
Vol.E93D
No.3
pp.647650 Publication Date: 2010/03/01 Online ISSN: 17451361
DOI: 10.1587/transinf.E93.D.647 Print ISSN: 09168532 Type of Manuscript: LETTER Category: Speech and Hearing Keyword: utterance verification, confidence measure, likelihood ratio testing, statelevel loglikelihood ratio, frame selection, state selection,
Full Text: PDF(180.1KB)>>
Summary:
This paper suggests utterance verification system using statelevel loglikelihood ratio with frame and state selection. We use hidden Markov models for speech recognition and utterance verification as acoustic models and antiphone models. The hidden Markov models have three states and each state represents different characteristics of a phone. Thus we propose an algorithm to compute statelevel loglikelihood ratio and give weights on states for obtaining more reliable confidence measure of recognized phones. Additionally, we propose a frame selection algorithm to compute confidence measure on frames including proper speech in the input speech. In general, phone segmentation information obtained from speakerindependent speech recognition system is not accurate because triphonebased acoustic models are difficult to effectively train for covering diverse pronunciation and coarticulation effect. So, it is more difficult to find the right matched states when obtaining state segmentation information. A state selection algorithm is suggested for finding valid states. The proposed method using statelevel loglikelihood ratio with frame and state selection shows that the relative reduction in equal error rate is 18.1% compared to the baseline system using simple phonelevel loglikelihood ratios.

