We’re running a video data collection study to record American Sign Language (ASL) and Black American Sign Language (BASL) videos. Participants will record short, voice-assistant-style commands (everyday phrases, common requests, conversational prompts) to help AI systems better understand signed requests.
This work directly supports more inclusive technology for Deaf and Hard-of-Hearing users. Each participant will complete 1,000 short videos, most around 10 seconds or less, recorded at home on an Apple device.
Record and annotate spontaneous Wolof speech from real-life conversations and AI interactions, providing recording metadata to help evaluate and improve speech recognition and conversational AI systems.
Record and annotate spontaneous Swati speech, capture approved conversations, and provide recording metadata to support the evaluation and improvement of speech recognition and conversational AI systems.
Record and annotate natural Southern Sotho conversations and speech interactions, providing audio and metadata that help evaluate and improve speech recognition and conversational AI systems.
Record and annotate spontaneous Slovak speech from everyday conversations and AI interactions, supplying recording metadata to support the evaluation and improvement of speech recognition and conversational AI systems.
Record and annotate spontaneous Norwegian Nynorsk speech in real-life conversations and AI interactions, providing metadata to support the evaluation and improvement of speech recognition and conversational AI systems.
