README.md

---
dataset_info:
- config_name: clean
  features:
  - name: utt_id
    dtype: string
  - name: speaker_id
    dtype: string
  - name: audio
    dtype: audio
  - name: transcript
    dtype: string
  - name: clean_transcription
    dtype: string
  - name: language
    dtype: string
  - name: annotated_transcription
    dtype: string
  splits:
  - name: train
    num_bytes: 13697637104.27
    num_examples: 48398
  - name: dev_man
    num_bytes: 1396705053.26
    num_examples: 4765
  - name: dev_sge
    num_bytes: 374969791.945
    num_examples: 1749
  download_size: 10638735107
  dataset_size: 15469311949.475
- config_name: default
  features:
  - name: utt_id
    dtype: string
  - name: speaker_id
    dtype: string
  - name: audio
    dtype: audio
  - name: transcript
    dtype: string
  splits:
  - name: train
    num_bytes: 13689096124.27
    num_examples: 48398
  - name: dev_man
    num_bytes: 1395851095.26
    num_examples: 4765
  - name: dev_sge
    num_bytes: 374671166.945
    num_examples: 1749
  download_size: 10632737228
  dataset_size: 15459618386.475
- config_name: phi4mm-fewshot-diffspeaker-whisper-prompt-v0-0-0
  features:
  - name: speaker_id
    dtype: string
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcript
    dtype: string
  - 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: dev_man
    num_bytes: 3421412819.36
    num_examples: 4628
  - name: dev_sge
    num_bytes: 936052966.925
    num_examples: 1735
  download_size: 3841413053
  dataset_size: 4357465786.285
- config_name: phi4mm-zeroshot-prompt-v0-0-0
  features:
  - name: speaker_id
    dtype: string
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcript
    dtype: string
  - name: prediction
    dtype: string
  splits:
  - name: dev_man
    num_bytes: 1387220593.784
    num_examples: 4628
  - name: dev_sge
    num_bytes: 373968270.265
    num_examples: 1735
  download_size: 1250324386
  dataset_size: 1761188864.0489998
configs:
- config_name: clean
  data_files:
  - split: train
    path: clean/train-*
  - split: dev_man
    path: clean/dev_man-*
  - split: dev_sge
    path: clean/dev_sge-*
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: dev_man
    path: data/dev_man-*
  - split: dev_sge
    path: data/dev_sge-*
- config_name: phi4mm-fewshot-diffspeaker-whisper-prompt-v0-0-0
  data_files:
  - split: dev_man
    path: phi4mm-fewshot-diffspeaker-whisper-prompt-v0-0-0/dev_man-*
  - split: dev_sge
    path: phi4mm-fewshot-diffspeaker-whisper-prompt-v0-0-0/dev_sge-*
- config_name: phi4mm-zeroshot-prompt-v0-0-0
  data_files:
  - split: dev_man
    path: phi4mm-zeroshot-prompt-v0-0-0/dev_man-*
  - split: dev_sge
    path: phi4mm-zeroshot-prompt-v0-0-0/dev_sge-*
---