Morphological Decomposition for Arabic Broadcast News Transcription
作者： Bing Xiang , Kham Nguyen , Long Nguyen , R. Schwartz , J. Makhoul
DOI: 10.1109/ICASSP.2006.1660214
关键词:
摘要: In this paper, we present a novel approach for morphological de-composition in large vocabulary Arabic speech recognition. It achieved low out-of-vocabulary (OOV) rate as well high recognition accuracy state-of-the-art broadcast news transcription system. approach, the compound words are decomposed into stems and affixes both language training acoustic data. The output re-joined before scoring. Four algorithms experimented compared work. best system 1.9% absolute reduction (9.8% relative) word error (WER) when to 64K-word baseline. performance of is also comparable 300K-word trained on normal words. meantime, much faster terms speed needs less memory than systems with larger 64K vocabularies.
sci-hub.se 参考文章(10)
Chafic Mokbel, Gérard Chollet, A. Ghaoui, François Yvon, On the use of morphological constraints in n-gram statistical language model. conference of the international speech communication association. pp. 1281- 1284 ,(2005)
John Makhoul, Mohamed Afify, Sherif M. Abdou, Bing Xiang, Long Nguyen, Recent progress in Arabic broadcast news transcription at BBN. conference of the international speech communication association. pp. 1637- 1640 ,(2005)
Gerhard Rigoll, Martha A. Larson, Joachim Köhler, Daniel Willett, Compound splitting and lexical unit recombination for improved performance of a speech recognition system for German parliamentary speeches. conference of the international speech communication association. pp. 945- 948 ,(2000)
Dimitra Vergyri, Andreas Stolcke, Kevin Duh, Katrin Kirchhoff, Morphology-Based Language Modeling for Arabic Speech Recognition conference of the international speech communication association. ,(2004)
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
K. Kirchhoff, J. Bilmes, S. Das, N. Duta, M. Egan, Gang Ji, Feng He, J. Henderson, Daben Liu, M. Noamany, P. Schone, R. Schwartz, D. Vergyri, Novel approaches to Arabic speech recognition: report from the 2002 Johns-Hopkins Summer Workshop international conference on acoustics, speech, and signal processing. ,vol. 1, pp. 344- 347 ,(2003) , 10.1109/ICASSP.2003.1198788
Roeland J.F. Ordelman, Franciska M.G. de Jong, Adrianus J. van Hessen, Compound Decomposition in Dutch Large Vocabulary Speech Recognition conference of the international speech communication association. ,(2003)
A. Berton, P. Fetter, P. Regel-Brietzmann, Compound words in large-vocabulary German speech recognition systems international conference on spoken language processing. ,vol. 2, pp. 1165- 1168 ,(1996) , 10.1109/ICSLP.1996.607814
Long Nguyen, Bing Xiang, Light supervision in acoustic model training international conference on acoustics, speech, and signal processing. ,vol. 1, pp. 185- 188 ,(2004) , 10.1109/ICASSP.2004.1325953
Kareem Darwish, Building a Shallow Arabic Morphological Analyser in One Day meeting of the association for computational linguistics. pp. 1- 8 ,(2002) , 10.3115/1118637.1118643