8th European Conference on Speech Communication and Technology

Geneva, Switzerland
September 1-4, 2003


A Dynamic Cross-Reference Pruning Strategy for Multiple Feature Fusion at Decoder Run Time

Yonghong Yan (1), Chengyi Zheng (1), Jianping Zhang (2), Jielin Pan (2), Jiang Han (2), Jian Liu (2)

(1) Oregon Health & Science University, USA
(2) Chinese Academy of Sciences, China

Although multiple cues, such as different signal processing techniques and feature representations, have been used in speech recognition in adverse acoustic environment, how to maximally utilize the benefit of these cues is largely unsolved. In this paper, a novel search strategy is proposed. During parallel decoding of different feature streams, the intermediate outputs are cross-referenced to reduce pruning errors. Experiment results show this method significantly improved recognition performance on a noisy large vocabulary continuous speech task.

Full Paper

Bibliographic reference.  Yan, Yonghong / Zheng, Chengyi / Zhang, Jianping / Pan, Jielin / Han, Jiang / Liu, Jian (2003): "A dynamic cross-reference pruning strategy for multiple feature fusion at decoder run time", In EUROSPEECH-2003, 1177-1180.