metadata
license: mit
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*
- config_name: finetune
data_files:
- split: train
path: finetune/train-*
- split: validation
path: finetune/validation-*
- split: test
path: finetune/test-*
- config_name: labeled_reliable
data_files:
- split: validation
path: labeled_reliable/validation-*
- split: test
path: labeled_reliable/test-*
- split: train
path: labeled_reliable/train-*
- config_name: phi4mm-zeroshot
data_files:
- split: validation
path: phi4mm-zeroshot/validation-*
- config_name: whisper-base-Min0_5-Max30_0-MinWordLen2
data_files:
- split: test
path: whisper-base-Min0_5-Max30_0-MinWordLen2/test-*
- config_name: whisper-base-en-Min0_5-Max30_0-MinWordLen2
data_files:
- split: test
path: whisper-base-en-Min0_5-Max30_0-MinWordLen2/test-*
- config_name: whisper-large-Min0_5-Max30_0-MinWordLen2
data_files:
- split: test
path: whisper-large-Min0_5-Max30_0-MinWordLen2/test-*
- config_name: whisper-large-v2_filtered_le50wer_lt30_ge3words
data_files:
- split: train
path: whisper-large-v2_filtered_le50wer_lt30_ge3words/train-*
- split: validation
path: whisper-large-v2_filtered_le50wer_lt30_ge3words/validation-*
- split: test
path: whisper-large-v2_filtered_le50wer_lt30_ge3words/test-*
- config_name: whisper-large-v3-Min0_5-Max30_0-MinWordLen0
data_files:
- split: test
path: whisper-large-v3-Min0_5-Max30_0-MinWordLen0/test-*
- config_name: whisper-large-v3-Min0_5-Max30_0-MinWordLen2
data_files:
- split: test
path: whisper-large-v3-Min0_5-Max30_0-MinWordLen2/test-*
- config_name: whisper-large-v3-turbo-Min0_5-Max30_0-MinWordLen2
data_files:
- split: test
path: whisper-large-v3-turbo-Min0_5-Max30_0-MinWordLen2/test-*
- config_name: whisper-large-v3-turbo-RSR_segmented-lora-Min0_5-Max30_0-MinWordLen0
data_files:
- split: test
path: >-
whisper-large-v3-turbo-RSR_segmented-lora-Min0_5-Max30_0-MinWordLen0/test-*
- config_name: whisper-largev3-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0
data_files:
- split: test
path: whisper-largev3-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0/test-*
- config_name: whisper-medium-Min0_5-Max30_0-MinWordLen2
data_files:
- split: test
path: whisper-medium-Min0_5-Max30_0-MinWordLen2/test-*
- config_name: whisper-medium-en-Min0_5-Max30_0-MinWordLen2
data_files:
- split: test
path: whisper-medium-en-Min0_5-Max30_0-MinWordLen2/test-*
- config_name: whisper-small-Min0_5-Max30_0-MinWordLen2
data_files:
- split: test
path: whisper-small-Min0_5-Max30_0-MinWordLen2/test-*
- config_name: whisper-small-en-Min0_5-Max30_0-MinWordLen2
data_files:
- split: test
path: whisper-small-en-Min0_5-Max30_0-MinWordLen2/test-*
- config_name: whisper-tiny-Min0_5-Max30_0-MinWordLen0
data_files:
- split: test
path: whisper-tiny-Min0_5-Max30_0-MinWordLen0/test-*
- config_name: whisper-tiny-Min0_5-Max30_0-MinWordLen2
data_files:
- split: test
path: whisper-tiny-Min0_5-Max30_0-MinWordLen2/test-*
- config_name: whisper-tiny-RSR_segmented-lora-Min0_5-Max30_0-MinWordLen0
data_files:
- split: test
path: whisper-tiny-RSR_segmented-lora-Min0_5-Max30_0-MinWordLen0/test-*
- config_name: whisper-tiny-en-Min0_5-Max30_0-MinWordLen2
data_files:
- split: test
path: whisper-tiny-en-Min0_5-Max30_0-MinWordLen2/test-*
- config_name: whisper-tiny-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0
data_files:
- split: test
path: whisper-tiny-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0/test-*
dataset_info:
- config_name: default
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
splits:
- name: train
num_bytes: 43732745924.625
num_examples: 181323
- name: validation
num_bytes: 5529107838.5
num_examples: 23652
- name: test
num_bytes: 5385316354
num_examples: 22592
download_size: 50216196525
dataset_size: 54647170117.125
- config_name: finetune
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
splits:
- name: train
num_bytes: 12603945060.375
num_examples: 51029
- name: validation
num_bytes: 2082115799.625
num_examples: 8459
- name: test
num_bytes: 2224927204
num_examples: 9200
download_size: 16055315812
dataset_size: 16910988064
- config_name: labeled_reliable
features:
- name: audio_path
dtype: audio
- name: duration
dtype: float64
- name: partition
dtype: string
- name: student_id
dtype: int64
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: float64
- name: part
dtype: int64
- name: transcription
dtype: string
- name: __index_level_0__
dtype: int64
- name: clean_transcription
dtype: string
splits:
- name: validation
num_bytes: 1305054292.577
num_examples: 9037
- name: test
num_bytes: 1625895604.264
num_examples: 10096
- name: train
num_bytes: 9046814046.34
num_examples: 55702
download_size: 10809717330
dataset_size: 11977763943.181
- config_name: phi4mm-zeroshot
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: duration
dtype: float64
- name: partition
dtype: string
- name: student_id
dtype: int64
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: float64
- name: part
dtype: int64
- name: transcription
dtype: string
- name: __index_level_0__
dtype: int64
- name: clean_transcription
dtype: string
- name: prediction
sequence: string
splits:
- name: validation
num_bytes: 2423050710.375
num_examples: 9037
download_size: 2300084662
dataset_size: 2423050710.375
- config_name: whisper-base-Min0_5-Max30_0-MinWordLen2
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2785747229.916
num_examples: 10834
download_size: 2646737982
dataset_size: 2785747229.916
- config_name: whisper-base-en-Min0_5-Max30_0-MinWordLen2
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2785727727.916
num_examples: 10834
download_size: 2646707554
dataset_size: 2785727727.916
- config_name: whisper-large-Min0_5-Max30_0-MinWordLen2
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2785738612.916
num_examples: 10834
download_size: 2646696617
dataset_size: 2785738612.916
- config_name: whisper-large-v2_filtered_le50wer_lt30_ge3words
features:
- name: utt_id
dtype: string
- name: audio
dtype: audio
- name: text
dtype: string
- name: duration
dtype: float64
splits:
- name: train
num_bytes: 8238959103.309
num_examples: 55647
- name: validation
num_bytes: 1123806544.375
num_examples: 9029
- name: test
num_bytes: 1430516243.515
num_examples: 10085
download_size: 10799753464
dataset_size: 10793281891.199
- config_name: whisper-large-v3-Min0_5-Max30_0-MinWordLen0
features:
- name: utt_id
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float64
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 1490300859
num_examples: 10076
download_size: 1421045142
dataset_size: 1490300859
- config_name: whisper-large-v3-Min0_5-Max30_0-MinWordLen2
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2785687969.916
num_examples: 10834
download_size: 2646685346
dataset_size: 2785687969.916
- config_name: whisper-large-v3-turbo-Min0_5-Max30_0-MinWordLen2
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2785695278.916
num_examples: 10834
download_size: 2646689389
dataset_size: 2785695278.916
- config_name: whisper-large-v3-turbo-RSR_segmented-lora-Min0_5-Max30_0-MinWordLen0
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2868668741.152
num_examples: 12348
download_size: 2718520625
dataset_size: 2868668741.152
- config_name: whisper-largev3-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0
features:
- name: utt_id
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float64
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 1490349324
num_examples: 10076
download_size: 1421074329
dataset_size: 1490349324
- config_name: whisper-medium-Min0_5-Max30_0-MinWordLen2
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2785717555.916
num_examples: 10834
download_size: 2646693573
dataset_size: 2785717555.916
- config_name: whisper-medium-en-Min0_5-Max30_0-MinWordLen2
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2785693899.916
num_examples: 10834
download_size: 2646688632
dataset_size: 2785693899.916
- config_name: whisper-small-Min0_5-Max30_0-MinWordLen2
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2785727432.916
num_examples: 10834
download_size: 2646701813
dataset_size: 2785727432.916
- config_name: whisper-small-en-Min0_5-Max30_0-MinWordLen2
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2785692360.916
num_examples: 10834
download_size: 2646693408
dataset_size: 2785692360.916
- config_name: whisper-tiny-Min0_5-Max30_0-MinWordLen0
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2868869550.152
num_examples: 12348
download_size: 2718697907
dataset_size: 2868869550.152
- config_name: whisper-tiny-Min0_5-Max30_0-MinWordLen2
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2785755651.916
num_examples: 10834
download_size: 2646735326
dataset_size: 2785755651.916
- config_name: whisper-tiny-RSR_segmented-lora-Min0_5-Max30_0-MinWordLen0
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2868383562.152
num_examples: 12348
download_size: 2718463502
dataset_size: 2868383562.152
- config_name: whisper-tiny-en-Min0_5-Max30_0-MinWordLen2
features:
- name: file
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float32
- name: student_id
dtype: string
- name: date
dtype: string
- name: time
dtype: string
- name: module
dtype: string
- name: investigation
dtype: string
- name: part
dtype: string
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 2785761833.916
num_examples: 10834
download_size: 2646742283
dataset_size: 2785761833.916
- config_name: whisper-tiny-myst-fullfinetune-Min0_5-Max30_0-MinWordLen0
features:
- name: utt_id
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: duration
dtype: float64
- name: normalized_transcription
dtype: string
- name: prediction
dtype: string
- name: normalized_prediction
dtype: string
- name: wer
dtype: float64
splits:
- name: test
num_bytes: 1490318785
num_examples: 10076
download_size: 1421037824
dataset_size: 1490318785
NOTE: This is not a public dataset, I couldn't share access.
There are two config for this dataset:
By default, is loading the raw dataset with basically everything of the dataset
There is another configuration is for finetune use, which is a preprocessed myst subset with
- cleaned transcription (in the format w2v2 output)
- duration, only 2~20 sec audio entries will be seen in this subset
- *local file name: only for reference use locally.
from datasets import load_dataset
myst = load_dataset("MagicLuke/MyST", split="test")
myst = load_dataset("MagicLuke/MyST", "finetune", split="test")