SpeechDx: A Multi-Task Benchmark for Clinical Speech AI
Quick Answer
SpeechDx introduces a comprehensive benchmark for clinical speech AI, encompassing 12 datasets and 27 tasks across various health conditions.
Quick Take
SpeechDx introduces a comprehensive benchmark for clinical speech AI, encompassing 12 datasets and 27 tasks across various health conditions. It evaluates generalization in clinical speech models, revealing that large-scale models outperform domain-specific ones, yet no current model generalizes effectively across the clinical spectrum.
Key Points
- SpeechDx spans 12 datasets and 27 tasks related to clinical speech AI.
- Tasks are structured by stages of speech production: conceptualization, formulation, and articulation.
- Large-scale speech models provide the strongest baselines in performance evaluations.
- Domain-specific models only enhance performance on closely matched tasks.
- No current model reliably generalizes across diverse clinical speech conditions.
Paper Resources
Article Content
From source RSS / original summaryarXiv:2606. 17339v1 Announce Type: new Abstract: Speech offers a uniquely informative window into health by simultaneously engaging neurological, motor, respiratory, and vocal systems. Current clinical speech AI methods have largely progressed through isolated condition-specific studies, making results difficult to compare and generalization difficult to assess. We introduce SpeechDx, a large-scale benchmark for clinical speech AI spanning 12 datasets and 27 tasks across diverse health conditions.
To enable evaluation across shared clinical mechanisms, SpeechDx structures tasks by the stage of speech production they disrupt: conceptualization, formulation, and articulation. The benchmark tests generalization by including tasks with limited labeled data and evaluating the same health condition across multiple datasets, distinguishing clinically meaningful patterns from dataset artefacts.
We systematically evaluate 12 state-of-the-art audio encoders across all tasks and under zero-shot cross-condition transfer. Results show that large-scale speech models represent the strongest overall baselines, domain-specific models improve performance only on closely matched tasks, and no current representation generalizes reliably across the clinical speech landscape. SpeechDx establishes a shared evaluation framework for tracking progress toward general-purpose clinical speech representations
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →Arbor: Tree Search as a Cognition Layer for Autonomous Agents
Arbor introduces a multi-agent framework utilizing structured tree search for optimizing LLM inference, achieving up to 193% throughput-latency improvement compared to vendor-optimized systems. It employs an Orchestrator and Critic agent for stability and coordination, demonstrating hardware-agnostic performance with minimal variance.