---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: normalized_text
    dtype: string
  - name: source
    dtype: string
  - name: tokenize_error
    dtype: 'null'
  - name: sr
    dtype: int64
  - name: codes_len
    dtype: int64
  - name: removed_frames
    dtype: int64
  - name: input_ids
    list: int32
  - name: labels
    list: int64
  - name: attention_mask
    list: int8
  splits:
  - name: train
    num_bytes: 5989413
    num_examples: 900
  download_size: 1930378
  dataset_size: 5989413
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---