---
dataset_info:
  features:
  - name: client_id
    dtype: string
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: sentence
    dtype: string
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 4815147523.5
    num_examples: 41604
  - name: val
    num_bytes: 1439996858.125
    num_examples: 10559
  download_size: 5660284094
  dataset_size: 6255144381.625
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: val
    path: data/val-*
---