Novel approaches to Arabic speech recognition: report from the 2002 Johns-Hopkins Summer Workshop

TLDR

Arabic is widely spoken yet speech recognition research, especially for dialectal varieties, remains limited compared to other languages. The study reports on a 2002 Johns Hopkins Summer Workshop project aimed at recognizing dialectal Arabic. The authors addressed vowel omission, morphological complexity, and dialect–formal discrepancies by developing automatic vowel restoration, morphology‑based language modeling, and incorporating out‑of‑corpus language model data. These techniques achieved significant word‑error‑rate reductions on the LDC Arabic CallHome task.

Abstract

Although Arabic is currently one of the most widely spoken languages in the world, there has been relatively little speech recognition research on Arabic compared to other languages. Moreover, most previous work has concentrated on the recognition of formal rather than dialectal Arabic. This paper reports on our project at the 2002 Johns Hopkins Summer Workshop, which focused on the recognition of dialectal Arabic. Three problems were addressed: (a) the lack of short vowels and other pronunciation information in Arabic texts; (b) the morphological complexity of Arabic; and (c) the discrepancies between dialectal and formal Arabic. We present novel approaches to automatic vowel restoration, morphology-based language modeling and the integration of out-of-corpus language model data, and report significant word error rate improvements on the LDC Arabic CallHome task.