analiza aiqbench
StreamHear addresses the critical challenge of domain shift in streaming automatic speech recognition by leveraging semi-supervised learning techniques to utilize abundant unlabeled data alongside costly labeled datasets. The method combines pseudo-labeling strategies with a dynamic-programming realignment step to improve word placement accuracy across diverse audio conditions, offering a practical advancement for deploying robust ASR models without extensive retraining on every new domain.
StreamHear: Domain-Adapted Pseudo-Labeling for Semi-Supervised Streaming Speech Recognition
arXiv:2608.13717v1 Announce Type: new Abstract: Streaming automatic speech recognition (ASR) underperforms on domain-shifted target audio, where labeled in-domain data is costly to prepare while unlabeled audio is abundant. We present StreamHear, a semi-supervised pipeline that adapts a pretrained streaming student by fine-tuning an offline transducer teacher on the labeled training set, generating pseudo-labels on the unlabeled portion, and fine-tuning the student on the mixture. We further introduce a prior-regularized dynamic-programming realignment step that fixes chunk-level word placement using an ASR-hypothesis anchor. Across four datasets spanning financial calls, prepared read speech, and phone-quality dialogue, StreamHear consistently outperforms supervised student fine-tuning and narrows the gap to the offline teacher.
Abstract:Streaming automatic speech recognition (ASR) underperforms on domain-shifted target audio, where labeled in-domain data is costly to prepare while unlabeled audio is abundant. We present StreamHear, a semi-supervised pipeline that adapts a pretrained streaming student by fine-tuning an offline transducer teacher on the labeled training set, generating pseudo-labels on the unlabeled portion, and fine-tuning the student on the mixture. We further introduce a prior-regularized dynamic-programming realignment step that fixes chunk-level word placement using an ASR-hypothesis anchor. Across four datasets spanning financial calls, prepared read speech, and phone-quality dialogue, StreamHear consistently outperforms supervised student fine-tuning and narrows the gap to the offline teacher.
Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
arXiv:2608.13717 [cs.CL]
(or arXiv:2608.13717v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2608.13717 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 · 202 cuvinte
Subiecte conexe
ReCalMatch:Reliability-Calibrated Semantic Guidance for Semi-Supervised Fine-Grained Recognition
arXiv · 25 sept. 2026
Flick: Few Labels Text Classification using K-Aware Intermediate Learning in Multi-Task Low-Resource Languages
arXiv · 28 iul. 2026
ESRVS: Extreme Semi-Supervised Retinal Vessel Segmentation with a Single Annotated Image
arXiv · 28 iul. 2026
SafeECGMatch: Calibration-Aware Joint Frequency and Time Space Semi-Supervised Learning for Open-Set ECG Classification
arXiv · 31 iul. 2026