SPARTA: Speaker Profiling for ARabic TAlk

SPARTA: Speaker Profiling for ARabic TAlk

This paper proposes a novel approach to an automatic estimation of three speaker traits from Arabic speech: gender, emotion, and dialect. After showing promising results on different text classification tasks, the multi-task learning (MTL) approach is used in this paper for Arabic speech classificat...

Full description

Bibliographic Details
Main Authors: Wael Farhan, Muhy Eddin Za'Ter, Qusai Abu Obaidah, Hisham Al Bataineh, Zyad Sober, Hussein Al Natsheh
Format: Article
Language:English
Published: FRUCT 2021-01-01
Series:Proceedings of the XXth Conference of Open Innovations Association FRUCT
Subjects:
speech recognition
dialect detection
gender detection
arabic speech
multi-task learning
natural language processing
Online Access:https://www.fruct.org/publications/fruct28/files/Far.pdf

Internet

https://www.fruct.org/publications/fruct28/files/Far.pdf

Similar Items