%Aigaion2 BibTeX export from Idiap Publications %Monday 30 December 2024 06:32:50 PM @TECHREPORT{aradilla:rr08-14, author = {Aradilla, Guillermo and Bourlard, Herv{\'{e}} and Magimai.-Doss, Mathew}, projects = {Idiap}, title = {Using KL-based Acoustic Models in a Large Vocabulary Recognition Task}, type = {Idiap-RR}, number = {Idiap-RR-14-2008}, year = {2008}, institution = {IDIAP}, abstract = {Posterior probabilities of sub-word units have been shown to be an effective front-end for ASR. However, attempts to model this type of features either do not benefit from modeling context-dependent phonemes, or use an inefficient distribution to estimate the state likelihood. This paper presents a novel acoustic model for posterior features that overcomes these limitations. The proposed model can be seen as a HMM where the score associated with each state is the KL divergence between a distribution characterizing the state and the posterior features from the test utterance. This KL-based acoustic model establishes a framework where other models for posterior features such as hybrid HMM/MLP and discrete HMM can be seen as particular cases. Experiments on the WSJ database show that the KL-based acoustic model can significantly outperform these latter approaches. Moreover, the proposed model can obtain comparable results to complex systems, such as HMM/GMM, using significantly fewer parameters.}, pdf = {https://publications.idiap.ch/attachments/reports/2008/aradilla-idiap-rr-08-14.pdf}, postscript = {ftp://ftp.idiap.ch/pub/reports/2008/aradilla-idiap-rr-08-14.ps.gz}, ipdmembership={speech}, }