Audio-FLAN: New Large-Scale Dataset for Unified Audio Understanding and Generation
A team of researchers has introduced Audio-FLAN, a large-scale instruction-tuning dataset comprising over 100 million instances across 80 tasks spanning speech, music, and sound domains. The dataset addresses a gap in AI research where audio understanding and generation have historically been treated as separate problems, and instruction tuning — proven effective in text and vision — has had limited application to audio. Audio-FLAN aims to enable large language models to handle both understanding and generation tasks in a zero-shot manner, potentially accelerating the development of unified audio-language models.
Researchers from multiple institutions have published Audio-FLAN, a large-scale instruction-tuning dataset designed to bridge the divide between audio understanding and audio generation in large language models (LLMs). The dataset covers 80 diverse tasks across speech, music, and sound, with over 100 million instances, making it one of the most comprehensive audio instruction-tuning resources to date. The motivation stems from recent advances in audio tokenization that have made integrating audio into LLMs more feasible, yet existing approaches typically treat understanding tasks — such as transcription and comprehension — separately from generation tasks like speech or music synthesis. Instruction tuning has already demonstrated strong generalization and zero-shot capabilities in text and vision domains, but its application to audio has remained largely unexplored due to the absence of suitable unified datasets. Audio-FLAN is intended to fill that gap, providing a foundation for models that can seamlessly switch between understanding and generating audio content without task-specific fine-tuning. The dataset has been made publicly available on HuggingFace and GitHub. The paper was first submitted in February 2025 and revised in June 2026.
What's missing
Details on data sourcing, licensing, and potential biases in the underlying audio corpora — which could affect model fairness and reproducibility — are not addressed in the available abstract.
What different sources said
- arXiv cs.AICenter
Audio-FLAN: An Instruction-Following Dataset for Unified Audio Understanding and Generation of Speech, Music, and Sound
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.