metadata

language:
  - hi
license: mit
size_categories:
  - 100K<n<1M
task_categories:
  - text-to-speech
  - automatic-speech-recognition
pretty_name: Rajasthani Hindi Speech Dataset
dataset_info:
  features:
    - name: audio
      dtype: audio
    - name: sentence
      dtype: string
  splits:
    - name: train
      num_bytes: 7425995581.812981
      num_examples: 422603
    - name: test
      num_bytes: 74991388.79801954
      num_examples: 4269
  download_size: 7504372330
  dataset_size: 7500986970.611
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
tags:
  - webdataset

Rajasthani Hindi Speech Dataset

This dataset consists of audio recordings of participants reading out stories in Rajasthani Hindi, one sentence at a time. They had 98 participants from Soda, Rajasthan. Each participant read 30 stories. In total, we have 426872 recordings in this dataset. They had roughly 58 male participants and 40 female participants.

Point to Note: While random sampling suggests that most users have to their best effort tried to accurately read out the sentences, we have not performed any quality analysis on the data. There could be errors in some of the recordings.

Dataset Sources

Link: Download
Curated By: Project Karya

Dataset Structure

Contains two headers: audio and sentence containing the Audio file and sentence respectively.