Release: EmoTa Dataset Loader v2.0.0

🚀 Release of the emota-loader Python package for EmoTa!

This version includes a clean, easy-to-use dataset loader for the EmoTa Dataset. It allows researchers and developers to load audio samples along with their corresponding transcript, emotion label, speaker information, and regional metadata.

Features

• 📂 Structured dataset loading from local directory
• 🧠 Rich metadata support (emotion, speaker, region, transcript)
• 🎧 Audio path resolution for each sample
• 🔍 Easy-to-use Python API: from emota_loader import EmoTaDataset

Installation

pip install emota-loader

Example Usage

from emota_loader import EmoTaDataset

dataset = EmoTaDataset(root_dir="./EmoTa").samples
sample = dataset[0]

print("Transcript:", sample.transcript)
print("Emotion:", sample.emotion)
print("Speaker Region:", sample.speaker_region)
print("Audio Path:", sample.audio_path)

📦 PyPI: https://pypi.org/project/emota-loader/

Tamil Speech Emotion Recognition (SER) Dataset Metadata and Resources

This release includes metadata, an overview of the Tamil Speech Emotion Recognition (SER) dataset (936 utterances), data access details, and licensing information. Designed to support SER research.