Datasets › CASA

CASA (Clinical annotations for automatic Stuttering Assessment)

Introduced by Ana Rita Valente et al. in Clinical Annotations for Automatic Stuttering Severity Assessment31 May 2025 archive 2025-07-28

Annotation Guidlines

Three speech and language pathologists, with experience ranging from 2 to 40 years, independently annotated and analyzed the audiovisual samples sourced from the Fluencybank Adults Who Stutter(AWS) dataset. The annotations were created using the ELAN tool.

For boundary Annotations, the annotators marked the onset and offset of each stuttering instance by dragging the mouse from the beginning to the end of the event, following the recommendations from the book TOCS: Test of childhood stuttering by Gillam et al. (2009). Specifically: (1) for repetitions, the stuttering onset is the point where the repeated speech begins, and the offset is where fluent speech resumes; (2) for prolongations or blocks, the onset is the start of the disfluent speech sound, with the offset marked by the next fluent sound.

After identifying each stuttering moment, the annotators classified the disfluency type based on the LBDL taxonomy, secondary behaviors using Riley’s (2009) SSI-4 classification, and the tension level according to a 0-3 scale, as proposed by the article Characteristics of stuttering-like disfluencies in Dutch-speaking children by Boey et al.

The annotations are categorized into 3 broad groups

Core
  1. SR: Syllable Repetition, indicating repeated movement of an entire syllable;
  2. ISR: Incomplete Syllable Repetition, repetition of parts of syllables;
  3. MUR: Multisyllable Unit Repetition, repetitions involving multiple syllables;
  4. P: sound Prolongation; and
  5. B: Blocks.
Secondary
  1. V: Verbal behaviors, including noisy breathing, whistling, sniffing, blowing, or clicking sounds;
  2. FG: Facial Grimaces, such as jaw jerking, tongue protrusion, lip pressing, jaw muscle tension, or eye movements;
  3. HM: Head Movements, including backward, forward, or turning motions; and
  4. ME: Movements of Extremities, such as arm and hand gestures, torso movements, or leg motions.
Tension levels
  • 0 – no tension
  • 1 – mild, subtle physical tension
  • 2 – moderate tension, noticeable and distracting
  • 3 – severe tension, highly distracting and produced with considerable effort

Benchmarks archive 2025-07-28

No leaderboard in the archive resolves to this dataset.

Papers archive 2025-07-28

No paper in the archive has a leaderboard row on this dataset; the archive counts 1 paper for it but never published that list.

Dataset loaders archive 2025-07-28

No loader listed in the archive.

Tasks archive 2025-07-28

No task tagged in the archive.

License archive 2025-07-28

No licence recorded in the archive. Absence here is not a statement about the dataset's terms.

Modalities archive 2025-07-28

No modality tagged.

Languages archive 2025-07-28

Variants archive 2025-07-28

  • CASA

1 variant name, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections