analiza aiqbench
This research addresses error propagation in cascaded ASR-LLM pipelines by introducing cause-aware detectors that distinguish between acoustic, linguistic, and deletion failures. By leveraging deep latent representations to enable targeted multi-turn clarifications, the approach significantly reduces word error rates and improves downstream task performance across diverse domains.
Proactive for Uncertainty: Cause-Aware Error Diagnosis and Interactive Clarification for Spoken Dialogue Systems
arXiv:2605.25404v2 Announce Type: replace Abstract: Cascaded Automatic Speech Recognition - Large Language Model (ASR-LLM) pipelines remain popular for industrial Spoken Dialogue Systems (SDS), primarily because their decoupled design ensures perceptual verifiability. However, cascaded systems suffer from error propagation, as transcription failures inevitably cascade to subsequent components, thereby degrading the final interaction quality. Although ASR confidence scores offer a simple filter for unreliable inputs, this approach is fundamentally limited because it typically fails to detect deletion errors or to distinguish between acoustic (inability to hear clearly) and linguistic (inability to understand) mismatches, both of which require targeted recovery strategies. In this paper, we propose a cause-aware error recovery paradigm that fundamentally rethinks robustness in SDS. Unlike traditional confidence filtering, we introduce a suite of small precision-focused detectors that exploit deep ASR latent representations to disentangle token-level errors into perception, comprehension, and deletion failures. This fine-grained diagnostic intelligence empowers the LLM to orchestrate targeted, multi-turn clarification strategies, effectively transforming ambiguous signals into seamless user interactions. Experimental results validate the precision of our approach, which more than doubles the recall on domain-shift errors (57.96% vs. 23.66%) compared to baselines. Crucially, this diagnostic precision yields up to a 31% reduction in WER and a 19% improvement on the downstream task across diverse accents, distortions, and domains.
Abstract:Cascaded Automatic Speech Recognition - Large Language Model (ASR-LLM) pipelines remain popular for industrial Spoken Dialogue Systems (SDS), primarily because their decoupled design ensures perceptual verifiability. However, cascaded systems suffer from error propagation, as transcription failures inevitably cascade to subsequent components, thereby degrading the final interaction quality. Although ASR confidence scores offer a simple filter for unreliable inputs, this approach is fundamentally limited because it typically fails to detect deletion errors or to distinguish between acoustic (inability to hear clearly) and linguistic (inability to understand) mismatches, both of which require targeted recovery strategies. In this paper, we propose a cause-aware error recovery paradigm that fundamentally rethinks robustness in SDS. Unlike traditional confidence filtering, we introduce a suite of small precision-focused detectors that exploit deep ASR latent representations to disentangle token-level errors into perception, comprehension, and deletion failures. This fine-grained diagnostic intelligence empowers the LLM to orchestrate targeted, multi-turn clarification strategies, effectively transforming ambiguous signals into seamless user interactions. Experimental results validate the precision of our approach, which more than doubles the recall on domain-shift errors (57.96% vs. 23.66%) compared to baselines. Crucially, this diagnostic precision yields up to a 31% reduction in WER and a 19% improvement on the downstream task across diverse accents, distortions, and domains.
Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
arXiv:2605.25404 [cs.CL]
(or arXiv:2605.25404v2 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2605.25404 Focus to learn more arXiv-issued DOI via DataCite
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 · 305 cuvinte
Subiecte conexe
An unexamined cause of the OpenAI Hugging Face hacking incident: its binary performance metric
LessWrong · 23 sept. 2026
Underwriting Superintelligence: Backing Agents you can Sue — Rune Kvist, AIUC
Latent Space · 16 sept. 2026
From Uncertainty to Clinical Risk: Severity-Aware Conformal Planning for Interactive Medical Diagnosis
arXiv · 1 sept. 2026
Towards Trustworthy AI for Glioma Diagnosis: A Task-Aware Evaluation of Uncertainty Quantification
arXiv · 1 oct. 2026