README.md

metadata

dataset_info:
  features:
    - name: Show
      dtype: string
    - name: EpId
      dtype: int64
    - name: ClipId
      dtype: int64
    - name: Start
      dtype: int64
    - name: Stop
      dtype: int64
    - name: Unsure
      dtype: bool
    - name: PoorAudioQuality
      dtype: bool
    - name: Prolongation
      dtype: bool
    - name: Block
      dtype: bool
    - name: SoundRep
      dtype: bool
    - name: WordRep
      dtype: bool
    - name: DifficultToUnderstand
      dtype: bool
    - name: Interjection
      dtype: bool
    - name: NoStutteredWords
      dtype: bool
    - name: NaturalPause
      dtype: bool
    - name: Music
      dtype: bool
    - name: NoSpeech
      dtype: bool
    - name: path
      dtype: string
    - name: audio
      dtype: audio
    - name: text
      dtype: string
    - name: annotated_text
      dtype: string
    - name: simple_annotated_text
      dtype: string
  splits:
    - name: train
      num_bytes: 354368511.712
      num_examples: 3684
  download_size: 334039625
  dataset_size: 354368511.712
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

HeStutters - Annotated Stuttering Events for 3.6k Audio Clips

Adapted from the Apple SEP-28k dataset.