Research foundation
SPMamba — 2-source to 3-source separation
Extended SPMamba from two-source to three-source speech separation, presented at UEC ASEAN Seminar and Workshop 2026.
Overview
Separating several overlapping sources is the same problem the stethoscope and the cough listener face: heart and lung, cough and speech. This study extended the SPMamba model from two to three simultaneous speakers and became the technical base for the project’s acoustic devices.
My part
- Adapted SPMamba from two-source to three-source separation.
- Presented the work at UEC ASEAN Seminar and Workshop 2026.
Status note Research summary. Paper and code are not yet public.