CONF Rasipuram_ICASSP_2011/IDIAP Integrating articulatory features using Kullback-Leibler divergence based acoustic model for phoneme recognition Rasipuram, Ramya Magimai-Doss, Mathew https://publications.idiap.ch/index.php/publications/showcite/Rasipuram_Idiap-RR-02-2011 Related documents Proceedings IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2011 5192 - 5195 10.1109/ICASSP.2011.5947527 doi In this paper, we propose a novel framework to integrate articulatory features (AFs) into HMM- based ASR system. This is achieved by using posterior probabilities of different AFs (estimated by multilayer perceptrons) directly as observation features in Kullback-Leibler divergence based HMM (KL-HMM) system. On the TIMIT phoneme recognition task, the proposed framework yields a phoneme recognition accuracy of 72.4% which is comparable to KL-HMM system using posterior probabilities of phonemes as features (72.7%). Furthermore, a best performance of 73.5% phoneme recognition accuracy is achieved by jointly modeling AF probabilities and phoneme probabilities as features. This shows the efficacy and flexibility of the proposed approach. REPORT Rasipuram_Idiap-RR-02-2011/IDIAP Integrating Articulatory Features using Kullback-Leibler Divergence based Acoustic Model for Phoneme Recognition Rasipuram, Ramya Magimai-Doss, Mathew articulatory features Automatic Speech Recognition Kullback-Leibler divergence based hidden Markov model multilayer perceptron phonemes posterior probabilities EXTERNAL https://publications.idiap.ch/attachments/reports/2010/Rasipuram_Idiap-RR-02-2011.pdf PUBLIC Idiap-RR-02-2011 2011 Idiap February 2011 In this paper, we propose a novel framework to integrate articulatory features (AFs) into HMM- based ASR system. This is achieved by using posterior probabilities of different AFs (estimated by multilayer perceptrons) directly as observation features in Kullback-Leibler divergence based HMM (KL-HMM) system. On the TIMIT phoneme recognition task, the proposed framework yields a phoneme recognition accuracy of 72.4% which is comparable to KL-HMM system using posterior probabilities of phonemes as features (72.7%). Furthermore, a best performance of 73.5% phoneme recognition accuracy is achieved by jointly modelling AF probabilities and phoneme probabilities as features. This shows the efficacy and flexibility of the proposed approach.