Other/Unspecific

This work creates a dataset of walking strides to understand differences in gait movements of people with Parkinson's disease and healthy adults. Young, old healthy participants and patients with Parkinson's disease walk on a level ground. Healthy participants walked on a circular path for 15 minutes, and patients with Parkinson's walked on a long hallway for 6 minutes, while they all wore a gait belt and 3D accelerometer.

A dataset of healthy and pathalogical voices are collected to encourage work in connected computing systems. Participants vocalized the vowel sound of the letter A for 5 seconds without any interruptions.

This work aims to capture participants’ experience of stuttering to buid a new instrument (OASES) for measuring the overall impact of stuttering. Each participant completed a reading task and an interview to learn more about the behaviors and affective/cognitive features of living with stuttering.

This work aims to capture children’s experience of stuttering to buid a new instrument (OASES) for measuring the overall impact of stuttering. Each child participant completed a grade-appropriate reading task and an interview to learn more about the behaviors and affective/cognitive features of living with stuttering.

3D pose estimates of people undergoing stroke rehabilitation and healthy controls is collected to help build an automated model that can predict the effectiveness of a rehabilitation exercise. Kinect sensor data of subjects performing tasks with a robotic stroke rehabilitation device and the joint data was recorded along with the common clinical assessment scores.

The dataset contains stuttering event annotations for automatic detection of stuttering speech events and speech recognition systems for people with atypical speech patterns. 28k annotated clips (23 hours) of speech were curated from public podcasts largely consisting of people who stutter interviewing other people who stutter.