Resources for Open-ended Response Correctness Assessment for Audio Question Answering
AI & ML interests
None defined yet.
Recent Activity
Papers
FLiP: Towards understanding and interpreting multimodal multilingual sentence embeddings
SE-DiCoW: Self-Enrolled Diarization-Conditioned Whisper
DiariZen is a speaker diarization toolkit driven by AudioZen and Pyannote 3.1.
This collection showcases DeCRED (Decoder-Centric Regularisation in Encoder-Decoder) for ASR.
Resources for Open-ended Response Correctness Assessment for Audio Question Answering
DiCoW (Diarization-Conditioned Whisper) is a collection of speaker-aware ASR models developed by BUT-FIT, extending OpenAI’s Whisper.
DiariZen is a speaker diarization toolkit driven by AudioZen and Pyannote 3.1.
This collection showcases DeCRED (Decoder-Centric Regularisation in Encoder-Decoder) for ASR.