Acoustic-phonetic Features For Refining The Explicit Speech Segmentation

This paper describes the refinement of the automatic speech segmentation into phones obtained via Hidden Markov Models (HMM). This refinement is based on acoustic-phonetic features associated to different phone classes. The proposed system was evaluated using both a small speaker dependent Brazilian Portuguese speech database and a speaker independent speech database (TIMIT). The refinement was applied to the boundaries obtained by just running the Viterbi's algorithm on the HMMs associated to the different utterances. Improvements of 30% and 13% were achieved in the percentage of segmentation errors below 20 ms for the speaker dependent and speaker independent databases respectively.318531856Vidal, E., Marzal, A., A review and new approaches for automatic segmentation of speech signals (1990) Signal Processing V: Theories and Applications, pp. 43-53. , L. Torres, E. Masgrau and M. A. Lagunas eds, Elsevier Science Publishers B. V, ppToledano, D., Gómez, L.A., Grande, L.V., Automatic Phonetic Segmentation (2003) IEEE Transactions on Speech and Audio Processing, 11 (6). , Novembervan Hemert, J.P., Automatic Segmentation of Speech (1991) IEEE Transactions on Signal Processing, 39 (4), pp. 1008-1012. , AprilSaiJayram, A.K.V., Ramasubramanian, V., Sreenivas, T.V., Robust parameters for automatic segmentation of speech (2002) Proceedings of Acoustics, Speech and Signals Processing, 1, pp. 513-512Hatazaki, K., Komor, Y., Kawabata, T., Shikano, K., Phoneme segmentation using spectrogram reading knowledge (1989) Proceeding of the International Conference on Acoustics Speech and Signal Processing, pp. 393-396Selmini, A. M. and Violaro, F. Improving the Explicit Automatic Speech Segmentation Provided by HMMs, Proceedings of the International Workshop on Telecommunications, pp. 220-226, Santa Rita do Sapucaí, Brazil, 2007Wang, L., Zhao, Y., Chu, M., Zhou, J., Cao, Z., Refining segmental boundaries for TTS database using fine contextual-dependent boundary models (2004) Proceedings of the ICASP, 1, pp. 641-644. , Beijing, China, MayDemuynck, K., Laureys, T., A comparison of different approaches to automatic speech segmentation (2002) Proceedings of the 5th International Conference on Text, Speech and Dialogue, , Brno, Czech Republic, SeptemberJuneja, A., Speech Recognition Based on Phonetic Features and Acoustic Landmarks (2004), PhD Thesis, University of Maryland, College Park, USAHosom, J.P., Automatic Phoneme Alignment Based on Acoustic-Phonetic Modeling (2002) 7th International Conference on Spoken Language Processing, , Denver, CO, USA, Septembe