DnR-nonverbal

홈페이지 · 논문 1편

DnR-nonverbal is a dataset for cinematic audio source separation (CASS) based on Divide and Remaster (DnR) dataset. Unlike conventional datasets, our dataset contains non-verbal sounds such as laughter and screaming, just like actual movie audio. Our dataset enables CASS models to allocate non-verbal sounds to the same stem as speech. Examples of clips and separation results are available at https://tky823.github.io/hasumi2025dnr.github.io/