Moustafa Elshafei (Keynote Speaker for CSP)

Application of deep learning in Arabic
speech processing and recognition

The presentation will give an overview of the recent advances in the applications of deep learning in language processing in general and speech recognition in particular. The talk will then provide a short review on the challenges facing Arabic Language and the recent research in Arabic language and Arabic speech. Finally we present our research activities along several Arabic language applications including, transcription of Arabic news, speech recognition of
Arabic dialects, Arabic language modeling using RNN, Arabic phoneme recognition, speech segmentation, automatic restoration of diacritical marks of Arabic text, Quran recitation, on-line Quran annotation, recognition of poem meters, and provide an overview of Arabic conversational agent.