DP-based wordgraph pruning

We present an efficient technique of generating word graphs in a continuous speech recognition system. The word graph is constructed in two stages. In the first stage, a huge word graph is generated as a by-product of a beam-driven forward search. Based on a dynamic-programming (DP) method, this huge word graph will be pruned in the second stage using higher level knowledge, such as n-gram language models. In this pruning stage an edge is removed if there is no path going through this edge which is better scored as the best-scored path in the word graph. The proposed technique is evaluated in the German VERBMOBIL task.

[1]  Mitch Weintraub,et al.  Large-vocabulary dictation using SRI's DECIPHER speech recognition system: progressive search techniques , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  Heinrich Niemann,et al.  Ergodic hidden Markov models and polygrams for language modeling , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.

[3]  Hermann Ney,et al.  Word graphs: an efficient interface between continuous-speech recognition and language understanding , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[4]  Wolfgang Wahlster,et al.  Verbmobil: Translation of Face-To-Face Dialogs , 1993, MTSUMMIT.

[5]  Peter Regel-Brietzmann,et al.  Evaluation of an HMM speech recognizer with various continuous speech databases , 1993, EUROSPEECH.

[6]  Peter Regel-Brietzmann,et al.  Optimization of an HMM - based continuous speech recognizer , 1993, EUROSPEECH.