From the 1 of 3 linked papers with an AI index.
3 papers
Investigating the Integration of Spatial Information in Foundation-Model-Based Speaker Diarization
Marc Deegen, Adrian Meise, Reinhold Haeb-Umbach
The paper evaluates how to combine spatial cues from multi‑channel audio with large pretrained single‑channel models for speaker diarization, finding that conditioning the diarizat…
Loose coupling of spectral and spatial models for multi-channel diarization and enhancement of meetings in dynamic environments
Adrian Meise, Tobias Cord-Landwehr, Christoph Boeddeker +3
Sound capture by microphone arrays opens the possibility to exploit spatial, in addition to spectral, information for diarization and signal enhancement, two important tasks in mee…
On the Application of Diffusion Models for Simultaneous Denoising and Dereverberation
Adrian Meise, Tobias Cord-Landwehr, Reinhold Haeb-Umbach
Diffusion models have been shown to achieve natural-sounding enhancement of speech degraded by noise or reverberation. However, their simultaneous denoising and dereverberation cap…