README.md

---
dataset_info:
  config_name: hi
  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: 641276980.0
    num_examples: 728
  - name: validation
    num_bytes: 362992514.0
    num_examples: 412
  - name: test
    num_bytes: 505581357.0
    num_examples: 584
  download_size: 1415114907
  dataset_size: 1509850851.0
configs:
- config_name: hi
  data_files:
  - split: train
    path: hi/train-*
  - split: validation
    path: hi/validation-*
  - split: test
    path: hi/test-*
---