---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 24000
  - name: sentence
    dtype: string
  - name: input_ids
    list:
      list: int32
  - name: attention_mask
    list:
      list: int8
  - name: labels
    list:
      list: int64
  splits:
  - name: train
    num_bytes: 316080285.0
    num_examples: 8000
  - name: test
    num_bytes: 92018065.0
    num_examples: 2000
  download_size: 372348543
  dataset_size: 408098350.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---