3 papers
eess.AS2026
Don't Listen to Me: A Lightweight, Low-Latency Model for Own-Voice Cancellation in Far-Field Speech Enhancement
Mads Ãstergaard, Alexander Neergaard Zahid, Karl Ulbæk +3
We introduce own-voice cancellation (OVC): removing a target (enrolled) speaker from a noisy multi-speaker mixture while preserving any remaining speech. Framed as the complement o…
cs.LG2026
Knowing When to Quit: Probabilistic Early Exits for Speech Separation
Kenny Falkær Olsen, Mads Ãstergaard, Karl Ulbæk +4
In recent years, deep learning-based single-channel speech separation has improved considerably, in large part driven by increasingly compute- and parameter-efficient neural networ…
cs.SD2024
SepMamba: State-space models for speaker separation using Mamba
Thor Højhus Avenstrup, Boldizsár Elek, István László Mádi +4
Deep learning-based single-channel speaker separation has improved significantly in recent years largely due to the introduction of the transformer-based attention mechanism. Howev…