---
dataset_info:
  features:
  - name: audio
    dtype: audio
  - name: speech_id
    dtype: string
  - name: speaker_id
    dtype: string
  - name: transcript
    dtype: string
  splits:
  - name: train
    num_bytes: 187936686
    num_examples: 1000
  download_size: 180001519
  dataset_size: 187936686
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
task_categories:
- automatic-speech-recognition
- text-to-speech
language:
- ml
pretty_name: Festvox IIITH Malayalam
size_categories:
- n<1K
---