metadata
dataset_info:
- config_name: CrisperWhisper
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371973404.598
num_examples: 2027
- name: train
num_bytes: 2487995781.375
num_examples: 14717
download_size: 2833340769
dataset_size: 2859969185.973
- config_name: CrisperWhisper-lora
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371971215.598
num_examples: 2027
download_size: 365041076
dataset_size: 371971215.598
- config_name: default
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
splits:
- name: test
num_bytes: 1035592320.849
num_examples: 2117
- name: train
num_bytes: 6896358826.488
num_examples: 15111
download_size: 7795518730
dataset_size: 7931951147.337
- config_name: hubert-large-ls960-ft
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: prediction
dtype: string
- name: clean_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
- name: normalized_transcription
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
splits:
- name: test
num_bytes: 374971742.971
num_examples: 2101
- name: train
num_bytes: 2500788472.33
num_examples: 15045
download_size: 2849221004
dataset_size: 2875760215.301
- config_name: hubert-xlarge-ls960-ft
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: prediction
dtype: string
- name: clean_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
- name: normalized_transcription
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
splits:
- name: test
num_bytes: 374963963.971
num_examples: 2101
- name: train
num_bytes: 2500728720.33
num_examples: 15045
download_size: 2849126054
dataset_size: 2875692684.301
- config_name: raw
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: normalized_transcription
dtype: string
splits:
- name: test
num_bytes: 1035591827.849
num_examples: 2117
- name: train
num_bytes: 6896354650.488
num_examples: 15111
download_size: 7795518600
dataset_size: 7931946478.337
- config_name: wav2vec2-base-960h
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: prediction
dtype: string
- name: clean_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
- name: normalized_transcription
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
splits:
- name: test
num_bytes: 374948193.971
num_examples: 2101
- name: train
num_bytes: 2500625127.33
num_examples: 15045
download_size: 2849437951
dataset_size: 2875573321.301
- config_name: wav2vec2-base-960h-lora
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
splits:
- name: test
num_bytes: 372246382
num_examples: 2032
download_size: 365622132
dataset_size: 372246382
- config_name: wav2vec2-conformer-rel-pos-large-960h-ft
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: prediction
dtype: string
- name: clean_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
- name: normalized_transcription
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
splits:
- name: test
num_bytes: 374973207.971
num_examples: 2101
- name: train
num_bytes: 2500804300.33
num_examples: 15045
download_size: 2849178918
dataset_size: 2875777508.301
- config_name: wav2vec2-conformer-rope-large-960h-ft
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: prediction
dtype: string
- name: clean_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
- name: normalized_transcription
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
splits:
- name: test
num_bytes: 374967250.971
num_examples: 2101
- name: train
num_bytes: 2500737023.33
num_examples: 15045
download_size: 2849079264
dataset_size: 2875704274.301
- config_name: wav2vec2-large-960h
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: prediction
dtype: string
- name: clean_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
- name: normalized_transcription
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
splits:
- name: test
num_bytes: 374971481.971
num_examples: 2101
- name: train
num_bytes: 2500788231.33
num_examples: 15045
download_size: 2849445778
dataset_size: 2875759713.301
- config_name: wav2vec2-large-robust-ft-libri-960h
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: prediction
dtype: string
- name: clean_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
- name: normalized_transcription
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
splits:
- name: test
num_bytes: 374964305.971
num_examples: 2101
- name: train
num_bytes: 2500724062.33
num_examples: 15045
download_size: 2849236158
dataset_size: 2875688368.301
- config_name: wav2vec2-xlsr-53-espeak-cv-ft
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: prediction
dtype: string
- name: clean_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
- name: normalized_transcription
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
splits:
- name: test
num_bytes: 375158787.971
num_examples: 2101
- name: train
num_bytes: 2502223226.33
num_examples: 15045
download_size: 2849866040
dataset_size: 2877382014.301
- config_name: whisper-base
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371967301.598
num_examples: 2027
- name: train
num_bytes: 2487952386.375
num_examples: 14717
download_size: 2833594374
dataset_size: 2859919687.973
- config_name: whisper-base-lora
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371973123.598
num_examples: 2027
download_size: 365043903
dataset_size: 371973123.598
- config_name: whisper-large
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371964939.598
num_examples: 2027
- name: train
num_bytes: 2487930450.375
num_examples: 14717
download_size: 2833269604
dataset_size: 2859895389.973
- config_name: whisper-large-v2
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371963949.598
num_examples: 2027
- name: train
num_bytes: 2487933397.375
num_examples: 14717
download_size: 2833248642
dataset_size: 2859897346.973
- config_name: whisper-large-v3
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371964151.598
num_examples: 2027
- name: train
num_bytes: 2487941624.375
num_examples: 14717
download_size: 2833238102
dataset_size: 2859905775.973
- config_name: whisper-large-v3-lora
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371962614.598
num_examples: 2027
download_size: 365029973
dataset_size: 371962614.598
- config_name: whisper-large-v3-turbo
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371964611.598
num_examples: 2027
- name: train
num_bytes: 2487942938.375
num_examples: 14717
download_size: 2833265201
dataset_size: 2859907549.973
- config_name: whisper-large-v3-turbo-lora
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371969121.598
num_examples: 2027
download_size: 365033961
dataset_size: 371969121.598
- config_name: whisper-medium
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371964274.598
num_examples: 2027
- name: train
num_bytes: 2487933757.375
num_examples: 14717
download_size: 2833292039
dataset_size: 2859898031.973
- config_name: whisper-medium-lora
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371971789.598
num_examples: 2027
download_size: 365034771
dataset_size: 371971789.598
- config_name: whisper-small
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371965060.598
num_examples: 2027
- name: train
num_bytes: 2487939380.375
num_examples: 14717
download_size: 2833395967
dataset_size: 2859904440.973
- config_name: whisper-small-lora
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371970732.598
num_examples: 2027
download_size: 365035549
dataset_size: 371970732.598
- config_name: whisper-tiny
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371976187.598
num_examples: 2027
- name: train
num_bytes: 2487967313.375
num_examples: 14717
download_size: 2833705989
dataset_size: 2859943500.973
- config_name: whisper-tiny-lora
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: string
- name: age
dtype: float64
- name: group
dtype: string
- name: split
dtype: string
- name: duration
dtype: float64
- name: clean_transcription
dtype: string
- name: prediction
dtype: string
- name: whisper_normalized_label
dtype: string
- name: whisper_normalized_prediction
dtype: string
- name: cer
dtype: float64
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 371976055.598
num_examples: 2027
download_size: 365050751
dataset_size: 371976055.598
configs:
- config_name: CrisperWhisper
data_files:
- split: test
path: CrisperWhisper/test-*
- split: train
path: CrisperWhisper/train-*
- config_name: CrisperWhisper-lora
data_files:
- split: test
path: CrisperWhisper-lora/test-*
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*
- config_name: hubert-large-ls960-ft
data_files:
- split: test
path: hubert-large-ls960-ft/test-*
- split: train
path: hubert-large-ls960-ft/train-*
- config_name: hubert-xlarge-ls960-ft
data_files:
- split: test
path: hubert-xlarge-ls960-ft/test-*
- split: train
path: hubert-xlarge-ls960-ft/train-*
- config_name: raw
data_files:
- split: test
path: raw/test-*
- split: train
path: raw/train-*
- config_name: wav2vec2-base-960h
data_files:
- split: test
path: wav2vec2-base-960h/test-*
- split: train
path: wav2vec2-base-960h/train-*
- config_name: wav2vec2-base-960h-lora
data_files:
- split: test
path: wav2vec2-base-960h-lora/test-*
- config_name: wav2vec2-conformer-rel-pos-large-960h-ft
data_files:
- split: test
path: wav2vec2-conformer-rel-pos-large-960h-ft/test-*
- split: train
path: wav2vec2-conformer-rel-pos-large-960h-ft/train-*
- config_name: wav2vec2-conformer-rope-large-960h-ft
data_files:
- split: test
path: wav2vec2-conformer-rope-large-960h-ft/test-*
- split: train
path: wav2vec2-conformer-rope-large-960h-ft/train-*
- config_name: wav2vec2-large-960h
data_files:
- split: test
path: wav2vec2-large-960h/test-*
- split: train
path: wav2vec2-large-960h/train-*
- config_name: wav2vec2-large-robust-ft-libri-960h
data_files:
- split: test
path: wav2vec2-large-robust-ft-libri-960h/test-*
- split: train
path: wav2vec2-large-robust-ft-libri-960h/train-*
- config_name: wav2vec2-xlsr-53-espeak-cv-ft
data_files:
- split: test
path: wav2vec2-xlsr-53-espeak-cv-ft/test-*
- split: train
path: wav2vec2-xlsr-53-espeak-cv-ft/train-*
- config_name: whisper-base
data_files:
- split: test
path: whisper-base/test-*
- split: train
path: whisper-base/train-*
- config_name: whisper-base-lora
data_files:
- split: test
path: whisper-base-lora/test-*
- config_name: whisper-large
data_files:
- split: test
path: whisper-large/test-*
- split: train
path: whisper-large/train-*
- config_name: whisper-large-v2
data_files:
- split: test
path: whisper-large-v2/test-*
- split: train
path: whisper-large-v2/train-*
- config_name: whisper-large-v3
data_files:
- split: test
path: whisper-large-v3/test-*
- split: train
path: whisper-large-v3/train-*
- config_name: whisper-large-v3-lora
data_files:
- split: test
path: whisper-large-v3-lora/test-*
- config_name: whisper-large-v3-turbo
data_files:
- split: test
path: whisper-large-v3-turbo/test-*
- split: train
path: whisper-large-v3-turbo/train-*
- config_name: whisper-large-v3-turbo-lora
data_files:
- split: test
path: whisper-large-v3-turbo-lora/test-*
- config_name: whisper-medium
data_files:
- split: test
path: whisper-medium/test-*
- split: train
path: whisper-medium/train-*
- config_name: whisper-medium-lora
data_files:
- split: test
path: whisper-medium-lora/test-*
- config_name: whisper-small
data_files:
- split: test
path: whisper-small/test-*
- split: train
path: whisper-small/train-*
- config_name: whisper-small-lora
data_files:
- split: test
path: whisper-small-lora/test-*
- config_name: whisper-tiny
data_files:
- split: test
path: whisper-tiny/test-*
- split: train
path: whisper-tiny/train-*
- config_name: whisper-tiny-lora
data_files:
- split: test
path: whisper-tiny-lora/test-*