Covertly improving intelligibility with data-driven adaptations of speech timing
arXiv:2603.30032v1 Announce Type: new Abstract: Human talkers often address listeners with language-comprehension challenges, such as hard-of-hearing or non-native adults, by globally slowing down their speech. However, it remains unclear whether this strategy actually makes speech more intelligible. Here, we take advantage of recent advancements in machine-generated speech allowing more precise control of speech rate in order to systematically examine how targeted speech-rate adjustments may improve comprehension. We first use reverse-correlation experiments to show that the temporal influence of speech rate prior to a target vowel contrast (ex. the tense-lax distinction) in fact manifests in a scissor-like pattern, with opposite effects in early versus late context windows; this pattern
View PDF HTML (experimental)
Abstract:Human talkers often address listeners with language-comprehension challenges, such as hard-of-hearing or non-native adults, by globally slowing down their speech. However, it remains unclear whether this strategy actually makes speech more intelligible. Here, we take advantage of recent advancements in machine-generated speech allowing more precise control of speech rate in order to systematically examine how targeted speech-rate adjustments may improve comprehension. We first use reverse-correlation experiments to show that the temporal influence of speech rate prior to a target vowel contrast (ex. the tense-lax distinction) in fact manifests in a scissor-like pattern, with opposite effects in early versus late context windows; this pattern is remarkably stable both within individuals and across native L1-English listeners and L2-English listeners with French, Mandarin, and Japanese L1s. Second, we show that this speech rate structure not only facilitates L2 listeners' comprehension of the target vowel contrast, but that native listeners also rely on this pattern in challenging acoustic conditions. Finally, we build a data-driven text-to-speech algorithm that replicates this temporal structure on novel speech sequences. Across a variety of sentences and vowel contrasts, listeners remained unaware that such targeted slowing improved word comprehension. Strikingly, participants instead judged the common strategy of global slowing as clearer, even though it actually increased comprehension errors. Together, these results show that targeted adjustments to speech rate significantly aid intelligibility under challenging conditions, while often going unnoticed. More generally, this paper provides a data-driven methodology to improve the accessibility of machine-generated speech which can be extended to other aspects of speech comprehension and a wide variety of listeners and environments.
Subjects:
Computation and Language (cs.CL); Sound (cs.SD)
Cite as: arXiv:2603.30032 [cs.CL]
(or arXiv:2603.30032v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2603.30032
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Paige Tuttösí [view email] [v1] Tue, 31 Mar 2026 17:30:47 UTC (1,666 KB)
Sign in to highlight and annotate this article
Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
announcejapanglobal
Looking for arXiv endorsement (cs.LG) – RL fine-tuning for VLMs (GRPO, MathVista)
Hi everyone, I am seeking an arXiv endorsement for cs.LG (Machine Learning) to submit my first paper on RL fine-tuning for vision-language models. Background: MS in AI (Purdue), working on RL + VLM training systems. Paper: A Case Study of Staged Metric-Gated GRPO for Visual Numeric Reasoning PDF: https://github.com/kgaero/RL_GSPO_Qwen2.5VLM/blob/main/paper/staged_metric_gated_grpo.pdf Short summary: Staged RL fine-tuning pipeline for VLMs (GRPO-based) Curriculum over MathVista subsets Metric-gated reward adaptation (structure → correctness) Checkpoint-aware continuation via alias-based selection Main result: Exact-match improves 0.375 → 0.75 with stable structure under constrained compute. If you’re eligible to endorse (cs.LG or related), I’d greatly appreciate it. Happy to share endorseme
The Programmer's Fulcrum: 03 April, 2026
Welcome to this week's The Programmer's Fulcrum. It's your weekly review of the essential news in the Open Media Network and Fediverse development communities with a focus on devastating big tech via Techno Anarchism. We aim to provide actionable content you can use to destroy Techno Feudalism each week. It has the additional benefit of weakening authoritarianism. IMHO, the best way to do that is to use tools from the Techno Anarchist Manifesto to build your own site(s) to participate in the Open Media Network . Then you should share it (them) via Real Simple Syndication (RSS), the Fediverse, and possibly a newsletter or podcast. This approach is similar to what some call the IndieWeb and its POSSE philosophy. The second best strategy is to have accounts on the Fediverse and use the hell o
Ask HN: Learning resources for building AI agents?
I’ve recently gone through several materials, including Antonio Gulli’s AI Agentic Design Patterns, Sam Bhagwat’s Principles of Building AI Agents and Patterns for Building AI Agents, as well as the courses from LangGraph Academy and some content on DataCamp. This space is evolving very quickly, so I’m curious how others here are approaching learning. What resources, courses, papers, or hands-on approaches have you found most useful while building AI agents? Comments URL: https://news.ycombinator.com/item?id=47637083 Points: 2 # Comments: 3
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
More in Countries
Priceless items are easy to steal. They're increasingly harder to sell.
Criminals are growing bolder, stealing priceless art , jewels and truckloads of goods — but it's harder than it looks for them to cash in on their heists. Why it matters: Because massive heists immediately dominate global news cycles, thieves quickly find themselves stuck with highly recognizable merchandise that even underground buyers are too afraid to touch. Driving the news: Thieves smashed into a small museum in the Italian countryside late last month, stealing three paintings worth over $10 million — a Renoir, a Cézanne and a Matisse. The operation took three minutes, and authorities are still investigating. The theft follows a similar heist last year at Paris' Louvre Museum, where thieves stole $104 million worth of France's crown jewels. Police arrested several suspects, but the na
Iran Threatens to Bomb 1GW Stargate AI Datacenter in Abu Dhabi
Article URL: https://timesofindia.indiatimes.com/technology/tech-news/iran-threatens-to-bomb-1gw-stargate-ai-datacenter-in-the-uae-shows-hidden-/articleshow/130003120.cms Comments URL: https://news.ycombinator.com/item?id=47637343 Points: 1 # Comments: 0
Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!