cantnlp@DravidianLangTech2025: A Bag-of-Sounds Approach to Multimodal Hate Speech Detection
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912276390346752 |
|---|---|
| author | Wong, Sidney Li, Andrew |
| author_facet | Wong, Sidney Li, Andrew |
| contents | This paper presents the systems and results for the Multimodal Social Media Data Analysis in Dravidian Languages (MSMDA-DL) shared task at the Fifth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages (DravidianLangTech-2025). We took a `bag-of-sounds' approach by training our hate speech detection system on the speech (audio) data using transformed Mel spectrogram measures. While our candidate model performed poorly on the test set, our approach offered promising results during training and development for Malayalam and Tamil. With sufficient and well-balanced training data, our results show that it is feasible to use both text and speech (audio) data in the development of multimodal hate speech detection systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_07862 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | cantnlp@DravidianLangTech2025: A Bag-of-Sounds Approach to Multimodal Hate Speech Detection Wong, Sidney Li, Andrew Computation and Language This paper presents the systems and results for the Multimodal Social Media Data Analysis in Dravidian Languages (MSMDA-DL) shared task at the Fifth Workshop on Speech, Vision, and Language Technologies for Dravidian Languages (DravidianLangTech-2025). We took a `bag-of-sounds' approach by training our hate speech detection system on the speech (audio) data using transformed Mel spectrogram measures. While our candidate model performed poorly on the test set, our approach offered promising results during training and development for Malayalam and Tamil. With sufficient and well-balanced training data, our results show that it is feasible to use both text and speech (audio) data in the development of multimodal hate speech detection systems. |
| title | cantnlp@DravidianLangTech2025: A Bag-of-Sounds Approach to Multimodal Hate Speech Detection |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2503.07862 |