---
dataset_info:
  features:
  - name: language
    dtype: string
  - name: speaker_id
    dtype: string
  - name: prompt_id
    dtype: string
  - name: prompt
    dtype: string
  - name: segment_id
    dtype: string
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: duration
    dtype: float64
  - name: raw_text
    dtype: string
  - name: iso_639_3
    dtype: string
  - name: glottocode
    dtype: string
  - name: iso_15924
    dtype: string
  splits:
  - name: train
    num_bytes: 11909146889
    num_examples: 12317
  - name: test
    num_bytes: 1323668285
    num_examples: 1369
  download_size: 13229977056
  dataset_size: 13232815174
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---