AI RESEARCH
CodecSep: Prompt-Driven Universal Sound Separation on Neural Audio Codec Latents
arXiv CS.LG
•
ArXi:2509.11717v5 Announce Type: replace-cross Text-guided sound separation enables flexible audio editing, assistive listening, and open-domain source extraction, but systems such as AudioSep remain too expensive for low-latency edge or codec-mediated deployment. Existing neural audio codec separators are efficient, yet largely restricted to fixed stems or closed taxonomies. We Across dnr-v2 and five open-domain benchmarks, CodecSep consistently improves over AudioSep in SI-SDR, remains competitive in ViSQOL, and achieves clear gains in human.