Advances in the CMU/Interact Arabic GALE Transcription System

This paper describes the CMU/InterACT effort in developing an Arabic Automatic Speech Recognition (ASR) system for broadcast news and conversations within the GALE 2006 evaluation. Through the span of 9 month in preparation for this evaluation we improved our system by 40% relative compared to our legacy system. These improvements have been achieved by various steps, such as developing a vowelized system, combining this system with a non-vowelized one, harvesting transcripts of TV shows from the web for slightly supervised training of acoustic models, as well as language model adaptation, and finally fine-tuning the overall ASR system.

[1]  Puming Zhan,et al.  Speaker normalization based on frequency warping , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  Bing Xiang,et al.  Light supervision in acoustic model training , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[3]  Tanja Schultz,et al.  Speaker segmentation and clustering in meetings , 2004, INTERSPEECH.

[4]  J. Xu,et al.  Audio Indexing of Arabic broadcast news , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[5]  Klaus Ries,et al.  The Karlsruhe-Verbmobil speech recognition engine , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[6]  Andreas Stolcke,et al.  SRILM - an extensible language modeling toolkit , 2002, INTERSPEECH.

[7]  Sherif Abdou,et al.  Recent progress in Arabic broadcast news transcription at BBN , 2005, INTERSPEECH.

[8]  A. Waibel,et al.  A one-pass decoder based on polymorphic linguistic context assignment , 2001, IEEE Workshop on Automatic Speech Recognition and Understanding, 2001. ASRU '01..

[9]  Tim Buckwalter.  Issues in Arabic Orthography and Morphology Analysis , 2004 .

[10]  Tanja Schultz,et al.  The ISL RT04 Mandarin Broadcast News Evaluation System , 2004 .