A real-time Japanese broadcast news closed-captioning system

This paper describes a collaboration between Bell Labs and NHK (Japan Broadcasting Corp.) STRL to develop a real-time large vocabulary speech recognition system for live closed-captioning of NHK news programs. Bell Labs broadcast news recognition engine consists of a two-pass decoder using bigram language models (LM) and right biphone models during the first pass, and trigram LM with within-word triphone models in the second pass. Various pruning strategies are used to achieve real time decoding, together with a noise compensation procedure aimed at improving recognition on noisy segments of the program. The system operates in a real-time mode and delivers less than 2% of word error rate (WER) on studio news conditions and about 5% of WER on noisy news and reporter speech when evaluated on a real broadcast news program.

[1]  Richard M. Schwartz,et al.  Single-tree method for grammar-directed search , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[2]  Kazuo Onoe,et al.  Time dependent language model for broadcast news transcription and its post-correction , 1998, ICSLP.

[3]  Olivier Siohan,et al.  Sequential noise estimation with optimal forgetting for robust speech recognition , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).

[4]  Wu Chou,et al.  Robust decision tree state tying for continuous speech recognition , 2000, IEEE Trans. Speech Audio Process..

[5]  R. A. Sukkar,et al.  Variable threshold vector quantization for reduced continuous density likelihood computation in speech recognition , 1997, 1997 IEEE Workshop on Automatic Speech Recognition and Understanding Proceedings.

[6]  Olivier Siohan,et al.  A NEW VERIFICATION-BASED FAST MATCH APPROACH TO LARGE VOCABULARY CONSTINUOUS SPEECH RECOGNITION , 2001 .

[7]  Shoei Sato,et al.  Progressive 2-pass decoder for real-time broadcast news captioning , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[8]  Qiru Zhou,et al.  An approach to continuous speech recognition based on layered self-adjusting decoding graph , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.