Abstract:This paper presents a unified multi-speaker encoder (UME), a novel architecture that jointly learns representations for speaker diarization (SD), speech separation (SS), and multi-speaker automatic speech recognition (ASR) tasks using a shared speech foundational encoder. We leverage the hidden representations from multiple layers of UME as a residual weighted-sum encoding (RWSE) to effectively use information from different semantic levels, contributing to bottom-up alignment between tasks. This joint training approach captures the inherent interdependencies among the tasks, enhancing overall performance on overlapping speech data. Our evaluations demonstrate that UME substantially improves over the single-task baselines dedicated to SD, SS, and multi-speaker ASR on LibriMix evaluation sets. Notably, for SD, UME outperforms the previous studies, achieving diarization error rates of 1.37% and 2.29% on Libri2Mix and Libri3Mix evaluation sets, respectively.
| Comments: | Accepted to IEEE ASRU 2025 |
| Subjects: | Audio and Speech Processing (eess.AS); Computation and Language (cs.CL); Sound (cs.SD) |
| Cite as: | arXiv:2508.20474 [eess.AS] |
| (or arXiv:2508.20474v2 [eess.AS] for this version) | |
| https://doi.org/10.48550/arXiv.2508.20474 arXiv-issued DOI via DataCite |
|
| Related DOI: | https://doi.org/10.1109/ASRU65441.2025.11434654
DOI(s) linking to related resources |
Submission history
From: Muhammad Shakeel [view email]
[v1]
Thu, 28 Aug 2025 06:50:57 UTC (3,937 KB)
[v2]
Wed, 13 May 2026 09:49:01 UTC (3,937 KB)