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-*