Compound words in large-vocabulary German speech recognition systems
作者： A. Berton , P. Fetter , P. Regel-Brietzmann
DOI: 10.1109/ICSLP.1996.607814
关键词:
摘要: The paper analyzes the impact of German compound words on speech recognition. It is well known that due to an idiosyncrasy orthography, make up a major fraction vocabulary, and most out-of-vocabulary (OOV) compounds are composed frequent already in lexicon. introduces new method for handling components rather than themselves. This not only reduces therefore complexity, but also improves word accuracy, reduced complexity means more robust language model.
isca-speech.org
udel.edu 参考文章(3)
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
P. Geutner, Using morphology towards better large-vocabulary speech recognition systems international conference on acoustics, speech, and signal processing. ,vol. 1, pp. 445- 448 ,(1995) , 10.1109/ICASSP.1995.479624
T. Kuhn, P. Fetter, A. Kaltenmeier, P. Regel-Brietzmann, DP-based wordgraph pruning international conference on acoustics speech and signal processing. ,vol. 2, pp. 861- 864 ,(1996) , 10.1109/ICASSP.1996.543257