Implementaion of Arabic Speech Emotion Recognition Employing Wav2vec2.0 and HuBERT Based on BAVED Dataset

Architecuture Model

Datasets:

name	Language	Access	Descreption
Basic Arabic Vocal Emotions Dataset	Arabic	Free ✔️	Basic Arabic Vocal Emotions Dataset (BAVED) is a datasetthat contains an arabic words spelled in diffrent levels of emotions recorded in an audio/wav format. The records contain 3 level of labeling , Level 0 is when the speaker is expressing a low level of emotion, this is similar to feeling tired or feeling down. Level 1 is the the standered level, it is the way the speaker speaks daily where he/she is expressing a neutral emotions,finally the level 2 emotion, its when the speaker is expressing a high level of positive or negative emotions (happiness, joy, sadness, anger, etc…).
Arabic Natural Audio Dataset	Arabic	Free ✔️	Eight videos of live calls between an anchor and a human outside the studio were downloaded from online Arabic talk shows. Each video was then divided into turns: callers and receivers. To label each video, 18 listeners were asked to listen to each video and select whether they perceive a happy, angry or surprised emotion. Silence, laughs and noisy chunks were removed. Every chunk was then automatically divided into 1 sec speech units forming our final corpus composed of 1384 records.
EYASE	Arabic	Free ✔️
KSUEmotions	Arabic	Paid	KSUEmotions was developed by King Saud University (KSU) and contains approximately five hours of emotional Modern Standard Arabic (MSA) speech from 23 subjects. Speakers were from three countries: Yemen, Saudi Arabia and Syria.

Models

Dataset	Model/Archit	Length(min)	No of records	Accuracy	Ckpt	ToDo
EYSAE	Elgeish / wav2vec2	19 min	499	64		✔️
EYSAE	othrif / wav2vec2	19 min	499	67		✔️
EYSAE	Zaid / Wave2vec2	19 min	499	67		✔️
EYSAE	hubert-base	19 min	499	70		✔️
EYSAE	hubert-large	19 min	499	64		✔️
BAVED	Elgeish	53 min	1935	89	ckpt	✔️
BAVED	hubert-base	53 min	1935	87		✔️
BAVED + EYSAE	Elgeish	7 2min	2434	84		✔️