CISAAD

Community Infrastructure to Strengthen AI for Audio Deepfake analysis (CISAAD) particularly for English language audio, is a prototype community resource focused on audio deepfake analysis. Developed by an interdisciplinary team across AI, linguistics, cyber infrastructure and human centered computing, this resource aims to address challenges related to information integrity, and increase awareness of audio deepfakes. The work will inform our understanding of mis/dis-information as a major societal concern and challenge, and also offer opportunities for content generation in positive applications such as voice restoration and smart and connected community research. It includes a deepfake data catalog and repository for English audio data, tools and models for deepfake audio analysis use cases, and educational resources.

Goals:
1) Address the challenges of limited data availability and human augmented data through creating a catalog of open datasets shared by the community,
2) Enable both single and multi-speaker deepfake analysis across various use cases, and
3) Address ethical, social, and political challenges associated with deploying deepfake technology developed from open-sourced community data.

 

NSF Award #2346473