5th European Conference on Speech Communication and Technology

Rhodes, Greece
September 22-25, 1997

Segment Boundary Estimation Using Recurrent Neural Networks

Toshiaki Fukada, Sophie Aveline, Mike Schuster, Yoshinori Sagisaka

ATR Interpreting Telecommunications Research Laboratories, Seika-cho, Soraku-gun, Kyoto, Japan

This paper describes a segment (e.g. phoneme) boundary estimation method based on recurrent neural networks (RNNs). The proposed method only requires acoustic observations to accurately estimate segment boundaries. Experimental results show that the proposed method can estimate segment boundaries significantly better than an HMM based method. Furthermore, we incorporate the RNN based segment boundary estimator into the HMM based and segment based recognition systems. As a result, the segment boundary estimates give useful information for reducing computational complexity and improving recognition performance.

Full Paper

Bibliographic reference.  Fukada, Toshiaki / Aveline, Sophie / Schuster, Mike / Sagisaka, Yoshinori (1997): "Segment boundary estimation using recurrent neural networks", In EUROSPEECH-1997, 2839-2842.