---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: path
    dtype: string
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcription
    dtype: string
  - name: duration
    dtype: float32
  - name: language
    dtype: string
  - name: original_speaker_id
    dtype: int64
  - name: session_id
    dtype: int64
  - name: topic
    dtype: string
  splits:
  - name: train
    num_bytes: 493354355.5925056
    num_examples: 4799
  - name: validation
    num_bytes: 61890319.82300885
    num_examples: 655
  - name: test
    num_bytes: 55305686.09885932
    num_examples: 685
  download_size: 496885228
  dataset_size: 610550361.5143738
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---