Improvements on automatic speech segmentation at the phonetic level

In this paper, we present some recent improvements in our automatic speech segmentation system, which only needs the speech signal and the phonetic sequence of each sentence of a corpus to be trained. It estimates a GMM by using all the sentences of the training subcorpus, where each Gaussian distribution represents an acoustic class, which probability densities are combined with a set of conditional probabilities in order to estimate the probability densities of the states of each phonetic unit. The initial values of the conditional probabilities are obtained by using a segmentation of each sentence assigning the same number of frames to each phonetic unit. A DTW algorithm fixes the phonetic boundaries using the known phonetic sequence. This DTW is a step inside an iterative process which aims to segment the corpus and re-estimate the conditional probabilities. The results presented here demonstrate that the system has a good capacity to learn how to identify the phonetic boundaries. © 2011 Springer-Verlag. This work was supported by the Spanish MICINN under contract TIN2008-06856-C05-02 Gómez Adrian, JA.; Calvo Lance, M. (2011). Improvements on automatic speech segmentation at the phonetic level. En Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. Springer Verlag (Germany). 7042:557-564. https://doi.org/10.1007/978-3-642-25085-9_66 Toledano, D.T., Hernández Gómez, L., Villarrubia Grande, L.: Automatic Phonetic Segmentation. IEEE Transactions on Speech and Audio Processing 11(6), 617–625 (2003) Kipp, A., Wesenick, M.B., Schiel, F.: Pronunciation modelling applied to automatic segmentation of spontaneous speech. In: Proceedings of Eurospeech, Rhodes, Greece, pp. 2013–2026 (1997) Sethy, A., Narayanan, S.: Refined Speech Segmentation for Concatenative Speech Synthesis. In: Proceedings of ICSLP, Denver, Colorado, USA, pp. 149–152 (2002) Jarify, S., Pastor, D., Rosec, O.: Cooperation between global and local methods for the automatic segmentation of speech synthesis corpora. In: Proceedings of Interspeech, Pittsburgh, Pennsylvania, USA, pp. 1666–1669 (2006) Romsdorfer, H., Pfister, B.: Phonetic Labeling and Segmentation of Mixed-Lingual Prosody Databases. In: Proceedings of Interspeech, Lisbon, Portual, pp. 3281–3284 (2005) Paulo, S., Oliveira, L.C.: DTW-based Phonetic Alignment Using Multiple Acoustic Features. In: Proceedings of Eurospeech, Geneva, Switzerland, pp. 309–312 (2003) Park, S.S., Shin, J.W., Kim, N.S.: Automatic