README.md

---
dataset_info:
  features:
  - name: path
    dtype: string
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: sentence
    dtype: string
  - name: condition_on_prev
    sequence: int64
  - name: whisper_transcript
    dtype: string
  splits:
  - name: train
    num_bytes: 631685248.0
    num_examples: 714
  - name: validation
    num_bytes: 359421903.0
    num_examples: 406
  - name: test
    num_bytes: 502350075.0
    num_examples: 576
  download_size: 1398110261
  dataset_size: 1493457226.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---