Spatio-spectral diarization of meetings by combining TDOA-based segmentation and speaker embedding-based clustering
Authors:
- Tobias Cord-LandwehrTCTobias Cord-Landwehr
- Tobias GburrekTGTobias Gburrek
- Marc DeegenMDMarc Deegen
- Reinhold Haeb-UmbachRHReinhold Haeb-Umbach
Preprint
Abstract
We propose a spatio-spectral, combined model-based and data-driven diarization pipeline consisting of TDOA-based segmentation followed by embedding-based clustering. The proposed system requires neither access to multi-channel training data nor prior knowledge about the number or placement of microphones. It works for both a compact microphone array and distributed microphones, with minor adjustments. Due to its superior handling of overlapping speech during segmentation, the proposed pipeline significantly outperforms the single-channel pyannote approach, both in a scenario with a compact microphone array and in a setup with distributed microphones. Additionally, we show that, unlike fully spatial diarization pipelines, the proposed system can correctly track speakers when they change positions.Keywords
All Related Versions
- Version 1, 2025-06-19, ArXiv
- Version 2, 2025-08-29, ArXiv
- Published version: , 5223.