5th International Conference on Spoken Language Processing

Sydney, Australia
November 30 - December 4, 1998

Language Independent and Language Adaptive Large Vocabulary Speech Recognition

Tanja Schultz (1), Alex Waibel (2)

(1) Interactive Systems Laboratories, Germany
(2) Interactive Systems Laboratories, USA

This paper describes the design of a multilingual speech recognizer using an LVCSR dictation database which has been collected under the project GlobalPhone. This project at the University of Karlsruhe investigates LVCSR systems in 15 languages of the world, namely Arabic, Chinese, Croatian, English, French, German, Italian, Japanese, Korean, Portuguese, Russian, Spanish, Swedish, Tamil, and Turkish. Based on a global phoneme set we built different multilingual speech recognition systems for five of the 15 languages. Context dependent phoneme models are created data-driven by introducing questions about languages and language groups to our polyphone clustering procedure. We apply the resulting multilingual models to unseen languages and present several recognition results in language independent and language adaptive setups. The results indicate that the method of parameter sharing should be decided depending on whether multilingual or crosslingual speech recognition is projected.

Full Paper

Bibliographic reference.  Schultz, Tanja / Waibel, Alex (1998): "Language independent and language adaptive large vocabulary speech recognition", In ICSLP-1998, paper 0577.