4th International Conference on Spoken Language Processing

Philadelphia, PA, USA
October 3-6, 1996

Speech Recognition Based on Acoustically Derived Segment Units

T. Fukada (1), M. Bacchiani (2), Kuldip K. Paliwal (3), Yoshinori Sagisaka (1)

(1) ATR Interpreting Telecommunications Research Laboratories, Japan
(2) Department of ECS Engineering, Boston University, USA
(3) School of ME, Griffith University, Australia

This paper describes a new method of word model generation based on acoustically derived segment units (henceforth ASUs). An ASU-based approach has the advantages of growing out of human pre-determined phonemes and of consistently generating acoustic units by using the maximum likelihood (ML) criterion. The former advantage is effective when it is difficult to map acoustics to a phone such as with highly co-articulated spontaneous speech. In order to implement an ASU-based modeling approach in a speech recognition system, we must first solve two points: (1) How do we design an inventory of acoustically-derived segmental units and (2) How do we model the pronunciations of lexical entries in terms of the ASUs. As for the second question, we propose an ASU-based word model generation method by composing the ASU statistics, that is, their means, variances and durations. The effectiveness of the proposed method is shown through spontaneous word recognition experiments.

Full Paper

Bibliographic reference.  Fukada, T. / Bacchiani, M. / Paliwal, Kuldip K. / Sagisaka, Yoshinori (1996): "Speech recognition based on acoustically derived segment units", In ICSLP-1996, 1077-1080.