DP-based wordgraph pruning
作者： T. Kuhn , P. Fetter , A. Kaltenmeier , P. Regel-Brietzmann
DOI: 10.1109/ICASSP.1996.543257
关键词: Word (computer architecture) 、 Speech recognition 、 Path (graph theory) 、 Language model 、 Word error rate 、 Rule-based machine translation 、 Graph theory 、 Speech processing 、 Graph 、 Pruning (morphology) 、 Natural language 、 Computer science
摘要: We present an efficient technique of generating word graphs in a continuous speech recognition system. The graph is constructed two stages. In the first stage, huge generated as by-product beam-driven forward search. Based on dynamic-programming (DP) method, this will be pruned second stage using higher level knowledge, such n-gram language models. pruning edge removed if there no path going through which better scored best-scored graph. proposed evaluated German VERBMOBIL task.
ieeecomputersociety.org参考文章(6)
Peter Regel-Brietzmann, Alfred Kaltenmeier, Fritz Class, Optimization of an HMM - based continuous speech recognizer. conference of the international speech communication association. ,(1993)
Peter Regel-Brietzmann, Alfred Kaltenmeier, Fritz Class, Evaluation of an HMM speech recognizer with various continuous speech databases. conference of the international speech communication association. ,(1993)
Wolfgang Wahlster, Verbmobil - Translation of Face-To-Face Dialogs Grundlagen und Anwendungen der Künstlichen Intelligenz, 17. Fachtagung für Künstliche Intelligenz, Humboldt-Universität zu. pp. 393- 402 ,(1993) , 10.22028/D291-24923
T. Kuhn, H. Niemann, E.G. Schukat-Talamazzini, Ergodic hidden Markov models and polygrams for language modeling international conference on acoustics, speech, and signal processing. pp. 357- 360 ,(1994) , 10.1109/ICASSP.1994.389282
H. Murveit, J. Butzberger, V. Digalakis, M. Weintraub, Large-vocabulary dictation using SRI's DECIPHER speech recognition system: progressive search techniques IEEE International Conference on Acoustics Speech and Signal Processing. ,vol. 2, pp. 319- 322 ,(1993) , 10.1109/ICASSP.1993.319301
M. Oerder, H. Ney, Word graphs: an efficient interface between continuous-speech recognition and language understanding IEEE International Conference on Acoustics Speech and Signal Processing. ,vol. 2, pp. 119- 122 ,(1993) , 10.1109/ICASSP.1993.319246