collaborators
Showing eess.ASShow all

12 papers · 1 filter

eess.AS2026

Diarization Error Decomposition Under Pause Annotation Ambiguity

Shota Horiguchi, Marc Delcroix, Naohiro Tawara +1

Speaker diarization evaluation is sensitive to ambiguity in pause annotation, which can inflate diarization error rate (DER) or obscure genuine model errors. We show that morpholog…

eess.AS2026

Over-Tightening-Aware Pseudo-Labeling for Tight-Boundary Speaker Diarization

Shota Horiguchi, Takanori Ashihara, Marc Delcroix +2

Training speaker diarization models on loose labels, such as speech segments with padded boundaries or filled pauses, often results in similarly loose model outputs. To obtain tigh…

eess.AS2026

Ontology-based Target Sound Extraction

Carlos Hernandez-Olivan, Marc Delcroix, Tsubasa Ochiai +2

Target sound extraction (TSE) aims to isolate a sound source of interest from a mixture, given a semantic query. Existing TSE systems are conditioned on fixed class representations…

eess.AS2026

SLT 2026 REAL-TSE Challenge: Real-world Target Speaker Extraction from Conversational Recordings

Shuai Wang, Zihan Qian, Ke Zhang +9

We introduce the REAL-TSE Challenge, an IEEE SLT 2026 satellite challenge on target speaker extraction~(TSE) from real conversational recordings. Given a multi-speaker mixture and…

eess.AS2026

SphereVBx: Spherical Variational Bayes Clustering for Simplified EEND-VC Diarization

Petr Pálka, Jiangyu Han, Prachi Singh +3

We propose SphereVBx, a Bayesian clustering framework for hyperspherical embeddings based on Toroidal Probabilistic Spherical Discriminant Analysis (T-PSDA). The method follows the…

eess.AS2026

Non-Autoregressive Minimum Bayes' Risk Decoding for Fast Speech Recognition

Hiroyuki Deguchi, Takatomo Kano, Katsuki Chousa +1

Non-autoregressive (NAR) decoding generates output tokens in parallel, making speech recognition faster than autoregressive decoding, which generates them sequentially from left to…