CMU_Kids / README.md
MagicLuke's picture
Upload dataset
49ce3c4 verified
metadata
license: mit
dataset_info:
  - config_name: checkpoint-1500-Min0_5-Max30_0-MinWordLen0
    features:
      - name: audio
        dtype:
          audio:
            sampling_rate: 16000
      - name: text
        dtype: string
      - name: duration
        dtype: float32
      - name: age
        dtype: string
      - name: speaker_id
        dtype: string
      - name: prompt
        dtype: string
      - name: location
        dtype: string
      - name: grade
        dtype: string
      - name: normalized_transcription
        dtype: string
      - name: prediction
        dtype: string
      - name: normalized_prediction
        dtype: string
      - name: wer
        dtype: float64
    splits:
      - name: train
        num_bytes: 1044176336
        num_examples: 5163
    download_size: 839764796
    dataset_size: 1044176336
  - config_name: checkpoint-3140-Min0_5-Max30_0-MinWordLen0
    features:
      - name: audio
        dtype:
          audio:
            sampling_rate: 16000
      - name: text
        dtype: string
      - name: duration
        dtype: float32
      - name: age
        dtype: string
      - name: speaker_id
        dtype: string
      - name: prompt
        dtype: string
      - name: location
        dtype: string
      - name: grade
        dtype: string
      - name: normalized_transcription
        dtype: string
      - name: prediction
        dtype: string
      - name: normalized_prediction
        dtype: string
      - name: wer
        dtype: float64
    splits:
      - name: train
        num_bytes: 1044311470
        num_examples: 5163
    download_size: 839787919
    dataset_size: 1044311470
  - config_name: clean
    features:
      - name: audio
        dtype:
          audio:
            sampling_rate: 16000
      - name: text
        dtype: string
      - name: duration
        dtype: float32
      - name: age
        dtype: string
      - name: speaker_id
        dtype: string
      - name: prompt
        dtype: string
      - name: location
        dtype: string
      - name: grade
        dtype: string
    splits:
      - name: train
        num_bytes: 956635618.6
        num_examples: 5180
    download_size: 847867606
    dataset_size: 956635618.6
  - config_name: default
    features:
      - name: audio
        dtype:
          audio:
            sampling_rate: 16000
      - name: text
        dtype: string
      - name: duration
        dtype: float32
      - name: age
        dtype: string
      - name: speaker_id
        dtype: string
      - name: prompt
        dtype: string
      - name: location
        dtype: string
      - name: grade
        dtype: string
    splits:
      - name: train
        num_bytes: 956824398.6
        num_examples: 5180
    download_size: 847957548
    dataset_size: 956824398.6
  - config_name: whisper-large-v3-Min0_5-Max30_0-MinWordLen0
    features:
      - name: audio
        dtype:
          audio:
            sampling_rate: 16000
      - name: text
        dtype: string
      - name: duration
        dtype: float32
      - name: age
        dtype: string
      - name: speaker_id
        dtype: string
      - name: prompt
        dtype: string
      - name: location
        dtype: string
      - name: grade
        dtype: string
      - name: normalized_transcription
        dtype: string
      - name: prediction
        dtype: string
      - name: normalized_prediction
        dtype: string
      - name: wer
        dtype: float64
    splits:
      - name: train
        num_bytes: 1044206457
        num_examples: 5163
    download_size: 839763689
    dataset_size: 1044206457
  - config_name: whisper-large-v3-turbo-Min0_5-Max30_0-MinWordLen0
    features:
      - name: audio
        dtype:
          audio:
            sampling_rate: 16000
      - name: text
        dtype: string
      - name: duration
        dtype: float32
      - name: age
        dtype: string
      - name: speaker_id
        dtype: string
      - name: prompt
        dtype: string
      - name: location
        dtype: string
      - name: grade
        dtype: string
      - name: normalized_transcription
        dtype: string
      - name: prediction
        dtype: string
      - name: normalized_prediction
        dtype: string
      - name: wer
        dtype: float64
    splits:
      - name: train
        num_bytes: 1044206392
        num_examples: 5163
    download_size: 839763524
    dataset_size: 1044206392
  - config_name: whisper-largev3-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0
    features:
      - name: audio
        dtype:
          audio:
            sampling_rate: 16000
      - name: text
        dtype: string
      - name: duration
        dtype: float32
      - name: age
        dtype: string
      - name: speaker_id
        dtype: string
      - name: prompt
        dtype: string
      - name: location
        dtype: string
      - name: grade
        dtype: string
      - name: normalized_transcription
        dtype: string
      - name: prediction
        dtype: string
      - name: normalized_prediction
        dtype: string
      - name: wer
        dtype: float64
    splits:
      - name: train
        num_bytes: 1044223554
        num_examples: 5163
    download_size: 839794737
    dataset_size: 1044223554
  - config_name: whisper-tiny-Min0_5-Max30_0-MinWordLen0
    features:
      - name: audio
        dtype:
          audio:
            sampling_rate: 16000
      - name: text
        dtype: string
      - name: duration
        dtype: float32
      - name: age
        dtype: string
      - name: speaker_id
        dtype: string
      - name: prompt
        dtype: string
      - name: location
        dtype: string
      - name: grade
        dtype: string
      - name: normalized_transcription
        dtype: string
      - name: prediction
        dtype: string
      - name: normalized_prediction
        dtype: string
      - name: wer
        dtype: float64
    splits:
      - name: train
        num_bytes: 1044228909
        num_examples: 5163
    download_size: 839790892
    dataset_size: 1044228909
  - config_name: whisper-tiny-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0
    features:
      - name: audio
        dtype:
          audio:
            sampling_rate: 16000
      - name: text
        dtype: string
      - name: duration
        dtype: float32
      - name: age
        dtype: string
      - name: speaker_id
        dtype: string
      - name: prompt
        dtype: string
      - name: location
        dtype: string
      - name: grade
        dtype: string
      - name: normalized_transcription
        dtype: string
      - name: prediction
        dtype: string
      - name: normalized_prediction
        dtype: string
      - name: wer
        dtype: float64
    splits:
      - name: train
        num_bytes: 1044208947
        num_examples: 5163
    download_size: 839777949
    dataset_size: 1044208947
configs:
  - config_name: checkpoint-1500-Min0_5-Max30_0-MinWordLen0
    data_files:
      - split: train
        path: checkpoint-1500-Min0_5-Max30_0-MinWordLen0/train-*
  - config_name: checkpoint-3140-Min0_5-Max30_0-MinWordLen0
    data_files:
      - split: train
        path: checkpoint-3140-Min0_5-Max30_0-MinWordLen0/train-*
  - config_name: clean
    data_files:
      - split: train
        path: clean/train-*
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
  - config_name: whisper-large-v3-Min0_5-Max30_0-MinWordLen0
    data_files:
      - split: train
        path: whisper-large-v3-Min0_5-Max30_0-MinWordLen0/train-*
  - config_name: whisper-large-v3-turbo-Min0_5-Max30_0-MinWordLen0
    data_files:
      - split: train
        path: whisper-large-v3-turbo-Min0_5-Max30_0-MinWordLen0/train-*
  - config_name: whisper-largev3-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0
    data_files:
      - split: train
        path: whisper-largev3-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0/train-*
  - config_name: whisper-tiny-Min0_5-Max30_0-MinWordLen0
    data_files:
      - split: train
        path: whisper-tiny-Min0_5-Max30_0-MinWordLen0/train-*
  - config_name: whisper-tiny-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0
    data_files:
      - split: train
        path: whisper-tiny-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0/train-*