Skip to content

Dataset · humyn-labs

Indian Emotional Speech: Audio and Text

humyn-labs/Indian-Emotional-Speech-Corpus

Audio recordings of Indian speakers across four emotions and paragraph text for emotion classification.

What's inside

The dataset contains high-quality audio recordings of Indian speakers reading a standard 50-word paragraph in four emotional tones: joy, sadness, surprise, and anger. Each recording lasts about 20-25 seconds and includes the full paragraph with tone shifts at specific points. The language is English, the size is under 1K records, and the modalities are audio and text.

What you can do with it

  • Fine-tuning an audio emotion classifier (e.g., wav2vec2)
  • Testing ASR on emotional English speech
  • Evaluating an emotion-aware voice agent
Tasks
Audio classification
Languages
English
Size
up to 1K rows
Downloads this month
251
Likes
441

How to load it

from datasets import load_dataset

ds = load_dataset("humyn-labs/Indian-Emotional-Speech-Corpus")