---
dataset_info:
  features:
  - name: file_name
    dtype: string
  - name: url
    dtype: string
  - name: audio_duration
    dtype: float64
  - name: original_id
    dtype: string
  - name: path
    dtype: string
  - name: strata
    dtype: string
  - name: age_group
    dtype: string
  - name: duration_category
    dtype: string
  - name: content_type
    dtype: string
  - name: uni
    dtype: string
  - name: base_transcript
    dtype: string
  - name: base_cer
    dtype: float64
  - name: ft_v1_inference_checkpoint-5000
    dtype: string
  - name: ft_v1_cer_5000
    dtype: float64
  - name: ft_v1_inference_checkpoint-10000
    dtype: string
  - name: ft_v1_cer_10000
    dtype: float64
  - name: ft_v1_inference_checkpoint-19000
    dtype: string
  - name: ft_v1_cer_19000
    dtype: float64
  - name: ft_v2_inference_checkpoint-20000
    dtype: string
  - name: ft_v2_cer_20000
    dtype: float64
  - name: ft_v2_inference_checkpoint-32000
    dtype: string
  - name: ft_v2_cer_32000
    dtype: float64
  - name: ft_v2_inference_checkpoint-43000
    dtype: string
  - name: ft_v2_cer_43000
    dtype: float64
  - name: ft_v3_inference_checkpoint-1000
    dtype: string
  - name: ft_v3_cer_1000
    dtype: float64
  - name: ft_v3_inference_checkpoint-25000
    dtype: string
  - name: ft_v3_cer_25000
    dtype: float64
  - name: ft_v4_inference_checkpoint-22000
    dtype: string
  - name: ft_v4_cer_22000
    dtype: float64
  splits:
  - name: train
    num_bytes: 2393414
    num_examples: 893
  download_size: 797013
  dataset_size: 2393414
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---