---
dataset_info:
  features:
  - name: 'No'
    dtype: int64
  - name: Song_id
    dtype: string
  - name: Chunk_id
    dtype: string
  - name: Title
    dtype: string
  - name: Reference
    dtype: string
  - name: Prediction
    dtype: string
  - name: WER (%) Case-Sensitive
    dtype: float64
  - name: WER (%) Lowercase
    dtype: float64
  - name: CER (%) Case-Sensitive
    dtype: float64
  - name: CER (%) Lowercase
    dtype: float64
  splits:
  - name: train
    num_bytes: 6168698
    num_examples: 15699
  download_size: 3163246
  dataset_size: 6168698
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---