EmoTa

Tamil Emotional Speech Dataset is a collection of recordings in Sri Lankan Tamil, representing the distinct dialects spoken in the northern, eastern, western, and central provinces. It aims to capture the linguistic and emotional diversity of these regions for use in speech and emotion recognition research.

---

---

EmoTa is the first emotional speech dataset in Tamil, designed to reflect the linguistic diversity of Sri Lankan Tamil speakers. It includes 936 utterances from 22 native Tamil speakers (11 male, 11 female), each articulating 19 semantically neutral sentences across five primary emotions: Anger, Happiness, Sadness, Fear, and Neutrality.

Key Features:

• Speakers: 22 native Tamil speakers (11 male, 11 female)
• Emotions: Anger, Happiness, Sadness, Fear, Neutrality
• Sentences: 19 semantically neutral sentences to reduce lexical bias
• Recording Quality: Captured in a controlled, soundproof environment with professional equipment
• Total Duration: Approx. 48 minutes of speech

Dataset Structure:

The dataset is organized into emotion-based folders with the following naming convention:

EmoTa/
    ├── happy/
    ├── sad/
    ├── angry/
    ├── fear/
    └── neutral/
        └── <spkID>_<senID>_<emo[:3]>.wav

Purpose:

EmoTa aims to facilitate research in Speech Emotion Recognition (SER) for the Tamil language, offering a balanced and diverse representation of emotional expressions from native Tamil speakers. It is released as open-access to support further exploration of Tamil language processing.

---

Citation

If you use EmoTa: A Tamil Emotional Speech Dataset in your research, please cite:

@inproceedings{thevakumar-etal-2025-emota,
    title = "{E}mo{T}a: A {T}amil Emotional Speech Dataset",
    author = "Thevakumar, Jubeerathan  and
      Thavarasa, Luxshan  and
      Sivatheepan, Thanikan  and
      Kugarajah, Sajeev  and
      Thayasivam, Uthayasanker",
    editor = "Sarveswaran, Kengatharaiyer  and
      Vaidya, Ashwini  and
      Krishna Bal, Bal  and
      Shams, Sana  and
      Thapa, Surendrabikram",
    booktitle = "Proceedings of the First Workshop on Challenges in Processing South Asian Languages (CHiPSAL 2025)",
    month = jan,
    year = "2025",
    address = "Abu Dhabi, UAE",
    publisher = "International Committee on Computational Linguistics",
    url = "https://aclanthology.org/2025.chipsal-1.19/",
    pages = "193--201",
    abstract = "This paper introduces EmoTa, the first emotional speech dataset in Tamil, designed to reflect the linguistic diversity of Sri Lankan Tamil speakers. EmoTa comprises 936 recorded utterances from 22 native Tamil speakers (11 male, 11 female), each articulating 19 semantically neutral sentences across five primary emotions: anger, happiness, sadness, fear, and neutrality. To ensure quality, inter-annotator agreement was assessed using Fleiss' Kappa, resulting in a substantial agreement score of 0.74. Initial evaluations using machine learning models, including XGBoost and Random Forest, yielded a high F1-score of 0.91 and 0.90 for emotion classification tasks. By releasing EmoTa, we aim to encourage further exploration of Tamil language processing and the development of innovative models for Tamil Speech Emotion Recognition."
}

Paper: view

---

Contact

🏷️ Name	📧 Email	🔗 LinkedIn
Jubeerathan Thevakumar	[email protected]	LinkedIn
Luxshan Thavarasa	[email protected]	LinkedIn
Thanikan Sivatheepan	[email protected]	LinkedIn
Uthayasanker Thayasivam	[email protected]	LinkedIn

Acknowledgment

Thanks to all the volunteers who provided samples to make EmoTa possible.
Special thanks to Braveenan Sritharan for his invaluable help and to Sajeev Kugarajah for his contribution to dataset collection.

---

🔗 Dataset Access