Researchers Adapt Voice Activity Prediction Model to Sign Language Interaction
A new study published on arXiv investigates whether Voice Activity Projection (VAP), a framework used to predict turn-taking in spoken conversation, can be adapted for sign language interaction with social robots. Using recordings from the Public DGS Corpus, researchers trained a model on hand, eye-region, and mouth-region pose features to predict signing activity. The work highlights both early promise and significant limitations, suggesting that sign-language-specific event definitions are needed before such systems can be reliably deployed.
Researchers have submitted a preprint to arXiv presenting an initial transfer study of Voice Activity Projection (VAP) models applied to dyadic sign language interaction, with the goal of enabling social robots to engage more inclusively with signing users. The study draws on interaction recordings from the Public DGS Corpus, deriving binary signing activity streams from lexical sign annotations to formulate proxy tasks for turn-taking prediction. The model incorporates pose-derived features from the hands, eye region, and mouth region of each signer. Results indicate that predicting whether a turn will be held or shifted (SHIFT/HOLD) is feasible, particularly when hand cues are used, but predicting the precise moment of a turn shift remains difficult. The authors conclude that while the VAP framework shows transferable potential, sign language interaction has structural properties that differ meaningfully from spoken interaction, requiring dedicated event definitions beyond those borrowed from speech research. The findings represent an early-stage contribution to a largely underexplored area of human-robot interaction accessibility.
What's missing
The study relies on binary signing activity streams derived from lexical sign annotations rather than full linguistic annotation, which may not capture the complexity of real sign language turn-taking (e.g., non-manual markers, spatial grammar). The corpus used (Public DGS Corpus) is specific to German Sign Language, so generalizability to other sign languages is untested. The paper is a preprint and has not yet undergone peer review. Baseline comparisons against chance or simple heuristic models are not described in the abstract, making it difficult to contextualize the reported performance levels.
What different sources said
- arXiv cs.CLCenter
Toward Signing Activity Projection in Sign Language Interaction
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.