Speech Technology

Nowadays there is a growing need for the integration of speech technology into the information technologies of today's information society. The acceptable development of automatic speech-to-text and text-to-speech technologies is a great challenge for the researchers in the field.

Research Area

My research objectives in the field of Speech Technology are:

- the development of a first-class Hungarian speech recognition system and speech synthesis
- the development of multimedia therapy systems with a speech interface for impaired learners
- the application of machine learning algorithms (mainly Kernel methods) to speech technology
- the development of novel signal processing methods

Selected Papers on the Topic

Bánhalmi, A., Kovács, K., Kocsor, A., Tóth, L.: Fundamental Frequency Estimation by Least-Squares Harmonic Model Fitting, Proceedings of INTERSPEECH, Lisbon, 2005. [pdf][abstract][bib]

Tóth, L., Kocsor, A., Gosztolya, G.: Telephone Speech Recognition via the Combination of Knowledge Sources in a Segmental Speech Model, Acta Cybernetica, Vol. 16, No. 4, 2004. [pdf][abstract][bib]

Tóth, L., Kocsor, A.: Harmonic Alternatives to Sine-Wave Speech, Proceedings of Euro-speech, Geneva, 2003. [pdf][abstract][bib]

Paczolay, D., Kocsor, A., Tóth, L.: Real-Time Vocal Tract Length Normalization in a Phonological Awareness Teaching System, in: Matousek, V., Mautner, P. (Eds.): Proceedings of TSD’2003, pp. 309-314, 2003. [pdf][abstract][bib]

Kocsor, A., Tóth, L., Kuba, A. Jr., Kovács, K., Jelasity, M., Gyimóthy, T., Csirik, J.: A Comparative Study of Several Feature Space Transformation and Learning Methods for Phoneme Classification, International Journal of Speech Technology, Vol. 3, No. 3/4, pp. 263-276, 2000. [pdf][abstract][bib]

Kocsor, A., Tóth, L., Bálint I.: On the Optimal Parameters of a Sinusoidal Representation of Signals, Acta Cybernetica, Vol. 14, No. 2, pp. 315-330, 1999. [pdf][abstract][bib]