---
dataset_info:
  features:
  - name: client_id
    dtype: string
  - name: path
    dtype: string
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcription
    dtype: string
  - name: up_votes
    dtype: int64
  - name: down_votes
    dtype: int64
  - name: age
    dtype: string
  - name: gender
    dtype: string
  - name: accent
    dtype: string
  - name: locale
    dtype: string
  - name: segment
    dtype: string
  - name: variant
    dtype: string
  - name: in_spectrogram
    sequence:
      sequence: float32
  splits:
  - name: train
    num_bytes: 6287282248.0
    num_examples: 20000
  - name: val
    num_bytes: 629444984.0
    num_examples: 2000
  download_size: 6825396882
  dataset_size: 6916727232.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: val
    path: data/val-*
---