King Saud University Emotions Corpus: Construction, Analysis, Evaluation, and Comparison
Emotional speech recognition for the Arabic language is insufficiently tackled in the literature compared to other languages. In this paper, we present the work of creating and verifying the King Saud University Emotions (KSUEmotions) corpus, which was released by the Linguistic Data Consortium (LDC...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9393909/ |
| _version_ | 1811210622108434432 |
|---|---|
| author | Ali Hamid Meftah Mustafa A. Qamhan Yasser Seddiq Yousef A. Alotaibi Sid Ahmed Selouani |
| author_facet | Ali Hamid Meftah Mustafa A. Qamhan Yasser Seddiq Yousef A. Alotaibi Sid Ahmed Selouani |
| author_sort | Ali Hamid Meftah |
| collection | DOAJ |
| description | Emotional speech recognition for the Arabic language is insufficiently tackled in the literature compared to other languages. In this paper, we present the work of creating and verifying the King Saud University Emotions (KSUEmotions) corpus, which was released by the Linguistic Data Consortium (LDC) in 2017 as the first public Arabic emotional speech corpus. KSUEmotions contains an emotional speech of twenty-three speakers from Saudi Arabia, Syria, and Yemen, and includes the emotions: neutral, happiness, sadness, surprise, and anger. The corpus content is verified in two different ways: a human perceptual test by nine listeners who rate emotional performance in audio files, and automatic emotion recognition. Two automatic emotion recognition systems are experimented with: Residual Neural Network and Convolutional Neural Network. This work also experiments with emotion recognition for the English language using the Emotional Prosody Speech and Transcripts Corpus (EPST). The current experimental work is conducted in three tracks: (i) monolingual, where independent experiments for Arabic and English are carried out, (ii) multilingual, where the Arabic and English corpora are merged in as mixed corpus, and (iii) cross-lingual, where models are trained using one language and tested using the other. A challenge encountered in this work is that the two corpora do not contain the same emotions. That problem is tackled by mapping the emotions to the arousal-valance space. |
| first_indexed | 2024-04-12T04:57:53Z |
| format | Article |
| id | doaj.art-4e7001295f4f48e9b43ad97994743a22 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-12T04:57:53Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-4e7001295f4f48e9b43ad97994743a222022-12-22T03:47:03ZengIEEEIEEE Access2169-35362021-01-019542015421910.1109/ACCESS.2021.30707519393909King Saud University Emotions Corpus: Construction, Analysis, Evaluation, and ComparisonAli Hamid Meftah0https://orcid.org/0000-0002-9038-2772Mustafa A. Qamhan1https://orcid.org/0000-0003-4463-4833Yasser Seddiq2https://orcid.org/0000-0001-7456-6478Yousef A. Alotaibi3https://orcid.org/0000-0003-0998-8978Sid Ahmed Selouani4https://orcid.org/0000-0003-0731-2632College of Computer and Information Sciences, King Saud University, Riyadh, Saudi ArabiaCollege of Computer and Information Sciences, King Saud University, Riyadh, Saudi ArabiaThe National Center for Electronics and Photonics Technology, King Abdulaziz City for Science and Technology (KACST), Riyadh, Saudi ArabiaCollege of Computer and Information Sciences, King Saud University, Riyadh, Saudi ArabiaDepartment of Information Managemen, Université de Moncton, Shippagan, NB, CanadaEmotional speech recognition for the Arabic language is insufficiently tackled in the literature compared to other languages. In this paper, we present the work of creating and verifying the King Saud University Emotions (KSUEmotions) corpus, which was released by the Linguistic Data Consortium (LDC) in 2017 as the first public Arabic emotional speech corpus. KSUEmotions contains an emotional speech of twenty-three speakers from Saudi Arabia, Syria, and Yemen, and includes the emotions: neutral, happiness, sadness, surprise, and anger. The corpus content is verified in two different ways: a human perceptual test by nine listeners who rate emotional performance in audio files, and automatic emotion recognition. Two automatic emotion recognition systems are experimented with: Residual Neural Network and Convolutional Neural Network. This work also experiments with emotion recognition for the English language using the Emotional Prosody Speech and Transcripts Corpus (EPST). The current experimental work is conducted in three tracks: (i) monolingual, where independent experiments for Arabic and English are carried out, (ii) multilingual, where the Arabic and English corpora are merged in as mixed corpus, and (iii) cross-lingual, where models are trained using one language and tested using the other. A challenge encountered in this work is that the two corpora do not contain the same emotions. That problem is tackled by mapping the emotions to the arousal-valance space.https://ieeexplore.ieee.org/document/9393909/Arabic languagecorpusCRNNdigital speech processingemotionResNet |
| spellingShingle | Ali Hamid Meftah Mustafa A. Qamhan Yasser Seddiq Yousef A. Alotaibi Sid Ahmed Selouani King Saud University Emotions Corpus: Construction, Analysis, Evaluation, and Comparison IEEE Access Arabic language corpus CRNN digital speech processing emotion ResNet |
| title | King Saud University Emotions Corpus: Construction, Analysis, Evaluation, and Comparison |
| title_full | King Saud University Emotions Corpus: Construction, Analysis, Evaluation, and Comparison |
| title_fullStr | King Saud University Emotions Corpus: Construction, Analysis, Evaluation, and Comparison |
| title_full_unstemmed | King Saud University Emotions Corpus: Construction, Analysis, Evaluation, and Comparison |
| title_short | King Saud University Emotions Corpus: Construction, Analysis, Evaluation, and Comparison |
| title_sort | king saud university emotions corpus construction analysis evaluation and comparison |
| topic | Arabic language corpus CRNN digital speech processing emotion ResNet |
| url | https://ieeexplore.ieee.org/document/9393909/ |
| work_keys_str_mv | AT alihamidmeftah kingsauduniversityemotionscorpusconstructionanalysisevaluationandcomparison AT mustafaaqamhan kingsauduniversityemotionscorpusconstructionanalysisevaluationandcomparison AT yasserseddiq kingsauduniversityemotionscorpusconstructionanalysisevaluationandcomparison AT yousefaalotaibi kingsauduniversityemotionscorpusconstructionanalysisevaluationandcomparison AT sidahmedselouani kingsauduniversityemotionscorpusconstructionanalysisevaluationandcomparison |