Interspeech'2005 - Eurospeech
The first part of this paper describes the BBN system that participated in the 2004 broadcast news (BN) evaluation for Arabic. The complete system description is given together with experimental results on the 2004 development, and evaluation sets. Previous Arabic speech recognition at BBN used grapheme models due to the lack of short vowel information in the acoustic transcriptions. In the second part of this paper we show how to build a phonetic system. It is demonstrated that switching to phonetic models is capable of reducing the word error rate by up to 14% relative, for different test sets, compared to the traditional grapheme based approach.
Bibliographic reference. Afify, Mohamed / Nguyen, Long / Xiang, Bing / Abdou, Sherif / Makhoul, John (2005): "Recent progress in Arabic broadcast news transcription at BBN", In INTERSPEECH-2005, 1637-1640.