metadata
dataset_info:
- config_name: add-clean-text-column
features:
- name: id
dtype: string
- name: path
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: duration
dtype: float32
- name: language
dtype: string
- name: original_speaker_id
dtype: int64
- name: session_id
dtype: int64
- name: topic
dtype: string
- name: phonetic_trans
dtype: string
- name: clean_transcription
dtype: string
splits:
- name: train
num_bytes: 1015994959.14
num_examples: 9869
- name: test
num_bytes: 106309996.135
num_examples: 1315
- name: validation
num_bytes: 106918680.43
num_examples: 1130
download_size: 1224406656
dataset_size: 1229223635.7050002
- config_name: default
features:
- name: id
dtype: string
- name: path
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: duration
dtype: float32
- name: language
dtype: string
- name: original_speaker_id
dtype: int64
- name: session_id
dtype: int64
- name: topic
dtype: string
- name: phonetic_trans
dtype: string
splits:
- name: train
num_bytes: 1015539069.14
num_examples: 9869
- name: test
num_bytes: 106265310.135
num_examples: 1315
- name: validation
num_bytes: 106871907.43
num_examples: 1130
download_size: 1223965462
dataset_size: 1228676286.7050002
- config_name: phi4mm-fewshot-diffspeaker-whisper-prompt-v0-0-0
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: original_speaker_id
dtype: int64
- name: audio_eg_1
dtype:
audio:
sampling_rate: 16000
- name: label_eg_1
dtype: string
- name: spkr_id_eg_1
dtype: string
- name: audio_eg_2
dtype:
audio:
sampling_rate: 16000
- name: label_eg_2
dtype: string
- name: spkr_id_eg_2
dtype: string
- name: audio_eg_3
dtype:
audio:
sampling_rate: 16000
- name: label_eg_3
dtype: string
- name: spkr_id_eg_3
dtype: string
- name: audio_eg_4
dtype:
audio:
sampling_rate: 16000
- name: label_eg_4
dtype: string
- name: spkr_id_eg_4
dtype: string
- name: prediction_w_1shots
dtype: string
- name: prediction_w_2shots
dtype: string
- name: prediction_w_3shots
dtype: string
- name: prediction_w_4shots
dtype: string
splits:
- name: test
num_bytes: 494116351
num_examples: 953
download_size: 436777692
dataset_size: 494116351
- config_name: phi4mm-zeroshot-prompt-v0-0-0
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: original_speaker_id
dtype: int64
- name: prediction
dtype: string
splits:
- name: test
num_bytes: 98836789
num_examples: 953
download_size: 98696614
dataset_size: 98836789
configs:
- config_name: add-clean-text-column
data_files:
- split: train
path: add-clean-text-column/train-*
- split: test
path: add-clean-text-column/test-*
- split: validation
path: add-clean-text-column/validation-*
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*
- config_name: phi4mm-fewshot-diffspeaker-whisper-prompt-v0-0-0
data_files:
- split: test
path: phi4mm-fewshot-diffspeaker-whisper-prompt-v0-0-0/test-*
- config_name: phi4mm-zeroshot-prompt-v0-0-0
data_files:
- split: test
path: phi4mm-zeroshot-prompt-v0-0-0/test-*