Automatic Phone Alignment: A Comparison between Speaker-Independent Models and Models Trained on the Corpus to Align

[en] Several automatic phonetic alignment tools have been proposed in the literature. They generally use speaker-independent acoustic models of the language to align new corpora. The problem is that the range of provided models is limited. It does not cover all languages and speaking styles (spontaneous, expressive, etc.). This study investigates the possibility of directly training the statistical model on the corpus to align. The main advantage is that it is applicable to any language and speaking style. Moreover, comparisons indicate that it provides as good or better results than using speaker-independent models of the language. It shows that about 2% are gained, with a 20 ms threshold, by using our method. Experiments were carried out on neutral and expressive corpora in French and English. The study also points out that even a small neutral corpus of a few minutes can be exploited to train a model that will provide high-quality alignment.

Disciplines :

Languages & linguistics

Author, co-author :

Brognaux, Sandrine

Roekhaut, Sophie

Drugman, Thomas

Beaufort, R.

Language :

English

Title :

Automatic Phone Alignment: A Comparison between Speaker-Independent Models and Models Trained on the Corpus to Align

Publication date :

19 October 2012

Event name :

International Conference on Natural Language Processing

Event place :

Kanazawa, Japan

Event date :

2012

Research unit :

F105 - Information, Signal et Intelligence artificielle

Research institute :

R450 - Institut NUMEDIART pour les Technologies des Arts Numériques

Available on ORBi UMONS :

since 19 October 2018

Statistics

Number of views

10 (0 by UMONS)

Number of downloads

0 (0 by UMONS)

More statistics

Scopus citations^®

Scopus citations^®
without self-citations

Bibliography

Kawai, H., Toda, T.: An evaluation of automatic phone segmentation for concatenative speech synthesis. In: Proc. of ICASSP 2004, Montreal, Canada, pp. 677-680 (2004)
Schiel, F., Draxler, C.: The production of speech corpora. Technical report, Bavarian Archive for Speech Signals (2003)
Goldman, J.P.: Easyalign: an automatic phonetic alignment tool under Praat. In: Proc. of Interspeech 2011, pp. 3233-3236 (2011)
Bigi, B., Hirst, D.: Speech phonetization alignment and syllabification (SPPAS): a tool for the automatic analysis of speech prosody. In: Proc. of Speech Prosody 2012 (2012)
Yuan, J., Liberman, M.: Speaker identification on the SCOTUS corpus. In: Proc. of Acoustics 2008, pp. 5687-5690 (2008)
Leggetter, C., Woodland, P.:Maximum likelihood linear regression for speaker adaptation of continuous density hidden Markov models. Computer Speech and Language 9(2), 171-185 (1995)
Adell, J., Bonafonte, A., Gomez, J.A., Castro, M.J.: Comparative study of automatic phone segmentation methods for TTS. In: Proc. of ICASSP 2005, pp. 309-312 (2005)
van Niekerk, D., Barnard, E.: Phonetic alignment for speech synthesis in under-resourced languages. In: Proc. of Interspeech 2009, Brighton, pp. 880-883 (2009)
Cangemi, F., Cutugno, F., Ludusan, B., Seppi, D., Van Compernolle, D.: Automatic speech segmentation for italian (ASSI): Tools, models, evaluation and applications. In: Proc. of AISV, Lecce, Italy, pp. 337-344 (2011)
Young, S., Kershaw, D., Odell, J., Ollason, D., Valtchev, V., Woodland, P.: The HTK Book (for HTK Version 3). Cambridge University (1995)
Lee, A., Kawahara, T., Shikano, K.: Julius-an open source real-time large vocabulary recognition engine. In: Proc. of Eurospeech 2001, pp. 1691-1694 (2001)
Toledano, D., Gomez, L.: HMMs for automatic phonetic segmentation. In: Proc. of LREC (2002)
Chen, L., Liu, Y., Harper, M., Maia, E., McRoy, S.: Evaluating factors impacting the accuracy of forced alignments in a multimodal corpus. In: Proc. of LREC 2004, pp. 759-762 (2004)
Ljolje, A., Hirschberg, J., van Santen, J.: Automatic speech segmentation for concatenative inventory selection. In: Second ESCA/IEEEWorkshop on Speech Synthesis, pp. 93-96 (1994)
Colotte, V., Beaufort, R.: Linguistic features weighting for a text-to-speech system without prosody model. In: Proc. of Interspeech 2005, pp. 2549-2552 (2005)
Dellaert, F., Polzin, T., Waibel, A.: Recognizing emotion in speech. In: Proc. of ICSLP, pp. 1970-1973 (1996)
Cosi, P., Falavigna, D., Omologo, M.: A preliminary statistical evaluation of manual and automatic segmentation discrepancies. In: Proc. of Eurospeech 1991, pp. 693-696 (1991)
MacLean, K.: VoxForge (2006-2012), http://www.voxforge.org