analiza aiqbench
This research addresses a critical bottleneck in cascaded multi-talker automatic speech recognition by introducing a pruning-based method to eliminate speaker leakage artifacts. By leveraging a pre-trained diarization model as a multimodal verifier, the approach achieves up to 29% relative error reduction in high-overlap scenarios without relying solely on lexical re-labeling. The technique significantly enhances transcript reliability for complex acoustic environments where traditional foundation models struggle with separation fidelity.
Mitigating Speaker Leakage in Cascaded Multi-talker ASR with Diarization-based Transcript Correction
arXiv:2608.22196v1 Announce Type: cross Abstract: While cascaded multi-talker ASR (MT-ASR) leverages state-of-the-art foundation models, its performance is often capped by speaker leakage during separation. Prior correction strategies primarily focus on lexical re-labeling for speaker attribution. We propose a complementary pruning-based paradigm that robustly identifies and removes leakage artifacts. Our method utilizes a pre-trained speaker diarization model as a multimodal verifier to prune transcribed segments satisfying a tripartite consensus of temporal containment, lexical cross-validation, and temporal alignment. Results on LibriMix, LibriSpeechMix, and the AMI Meeting corpus show our algorithm consistently reduces cpW ER across diverse overlap conditions. Specifically, on subsets with high speaker leakage, our method achieves relative cpW ER reductions of up to 29%, highlighting its effectiveness in enhancing the reliability of cascaded MT-ASR transcripts in complex acoustic environments.
Abstract:While cascaded multi-talker ASR (MT-ASR) leverages state-of-the-art foundation models, its performance is often capped by speaker leakage during separation. Prior correction strategies primarily focus on lexical re-labeling for speaker attribution. We propose a complementary pruning-based paradigm that robustly identifies and removes leakage artifacts. Our method utilizes a pre-trained speaker diarization model as a multimodal verifier to prune transcribed segments satisfying a tripartite consensus of temporal containment, lexical cross-validation, and temporal alignment. Results on LibriMix, LibriSpeechMix, and the AMI Meeting corpus show our algorithm consistently reduces cpW ER across diverse overlap conditions. Specifically, on subsets with high speaker leakage, our method achieves relative cpW ER reductions of up to 29%, highlighting its effectiveness in enhancing the reliability of cascaded MT-ASR transcripts in complex acoustic environments.
Audio and Speech Processing (eess.AS); Computation and Language (cs.CL)
arXiv:2608.22196 [eess.AS]
(or arXiv:2608.22196v1 [eess.AS] for this version)
https://doi.org/10.48550/arXiv.2608.22196 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 224 cuvinte
Subiecte conexe
Self-Supervised Speech Representations for Cross-Speaker Dysarthria Detection During Awake Craniotomy
arXiv · 9 oct. 2026
NVIDIA Releases Nemotron 3 Diarization: A 100M-Parameter Open-Weight Model That Tracks 8 Speakers in Real Time
MarkTechPost · 23 sept. 2026
On Low-Bit Quantization Errors in Speaker Verification: Diagnostic and Mitigation
arXiv · 9 iun. 2026
Open Distillation of Hereditary Traits
AI Alignment Forum · 14 iul. 2026