aiqbench analysis
ANCHOR addresses the critical challenge of incremental speech quality estimation in streaming and generative systems by reformulating the task as a multi-resolution autoregressive process. The method achieves a 48% reduction in PLCMOS error on short prefixes through a dual-resolution token architecture that enables coarse-to-fine refinement within a single decoder. This approach provides a robust framework for real-time quality monitoring where full audio context is unavailable.
ANCHOR: Autoregressive Non-intrusive Chunk-Ordered Refinement for Joint Multi-Resolution Speech Quality Modeling
arXiv:2606.10233v1 Announce Type: cross Abstract: While speech quality is typically assessed on complete utterances, streaming and generative systems require incremental estimation from partial audio. Existing predictors assume full context, degrading on prefix-constrained inputs. Extending ARECHO, we propose ANCHOR, reformulating incremental assessment as a multi-resolution autoregressive task. It models chunk- and utterance-level quality within a single decoder using dual-resolution tokens and a resolution-aware hierarchy for coarse-to-fine refinement. Experiments show substantial robustness under partial input, including a 48% PLCMOS error reduction on 2-second prefixes. Convergence analysis reveals a 4-6 s effective perceptual context horizon. A stress test further isolates structured extrapolation biases under localized corruption. Results demonstrate that hierarchical supervision improves incremental prediction and elucidates how perceptual quality accumulates over time.
Abstract:While speech quality is typically assessed on complete utterances, streaming and generative systems require incremental estimation from partial audio. Existing predictors assume full context, degrading on prefix-constrained inputs. Extending ARECHO, we propose ANCHOR, reformulating incremental assessment as a multi-resolution autoregressive task. It models chunk- and utterance-level quality within a single decoder using dual-resolution tokens and a resolution-aware hierarchy for coarse-to-fine refinement. Experiments show substantial robustness under partial input, including a 48% PLCMOS error reduction on 2-second prefixes. Convergence analysis reveals a 4-6 s effective perceptual context horizon. A stress test further isolates structured extrapolation biases under localized corruption. Results demonstrate that hierarchical supervision improves incremental prediction and elucidates how perceptual quality accumulates over time.
Audio and Speech Processing (eess.AS); Machine Learning (cs.LG); Sound (cs.SD)
arXiv:2606.10233 [eess.AS]
(or arXiv:2606.10233v1 [eess.AS] for this version)
https://doi.org/10.48550/arXiv.2606.10233 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.
Full text under the publisher's open licence · 215 words
Related coverage
Build a Physical AI model factory with NVIDIA Cosmos 3 on SageMaker HyperPod
AWS · Sep 4, 2026
MMaDA-VLA: Large Diffusion Vision-Language-Action Model with Unified Multi-Modal Instruction and Generation
arXiv · Aug 7, 2026
Revisiting Structural Dependency in Autoregressive Multi-Task Table Recognition via Order-Independent Cell-Level Representations
arXiv · Jun 17, 2026
Research Note: Filtering Subversion-Relevant Information From Pretraining Data Is Feasible
LessWrong · Oct 5, 2026