| Pushing the Limits of End-to-End Diarization | | 14.49 | EEND-TA + FT | 2025-09-18 |
| Efficient and Generalizable Speaker Diarization via Structured Pruning of Self-Supervised Models | ✓ Link | 14.5 | diarizen-wavlm-large-s80-md-v2 | 2025-06-23 |
| VBx for End-to-End Neural and Clustering-based Diarization | ✓ Link | 14.6 | DiariZen-Large-s80 + cVBx | 2025-10-22 |
| Speaker Diarization with Overlapping Community Detection Using Graph Attention Networks and Label Propagation Algorithm | | 15.94 | OCDGALP (ResNet + GAT + LPA-OCD) (extra data) | 2025-06-03 |
| End-to-end speaker segmentation for overlap-aware resegmentation | ✓ Link | 19.3 | VBx + pyannote 2.0 VAD and resegmentation | 2021-04-08 |
| Encoder-Decoder Based Attractors for End-to-End Neural Diarization | ✓ Link | 19.53 | EEND-EDA (iterative inference+, Hitachi-JHU SAD) | 2021-06-20 |
| DiaPer: End-to-End Neural Diarization with Perceiver-Based Attractors | ✓ Link | 21.1 | DiaPer 20-attractor + compound FT 5min | 2023-12-07 |
| Mitigating Non-Target Speaker Bias in Guided Speaker Embedding | | 21.18 | Bias-mitigated guided CAM++ (P2-1) + pyannote 3.1 | 2025-06-14 |
| Powerset multi-class cross entropy loss for neural speaker diarization | ✓ Link | 21.3 | Powerset (domain-adapted) | 2023-10-19 |
| The Third DIHARD Diarization Challenge | ✓ Link | 25.36 | DIHARD III Baseline (x-vector + VB-HMM) (extra data) | 2020-12-02 |