FormalASR: New AI Models Convert Spoken Chinese Directly to Formal Written Text
Researchers have introduced FormalASR, two compact end-to-end models (0.6B and 1.7B parameters) that transcribe spoken Chinese directly into formal written text without requiring a separate post-processing language model. Current ASR systems typically preserve disfluencies and informal speech patterns, and the common fix—chaining an ASR system with a large language model—adds latency and memory overhead that makes on-device deployment difficult. FormalASR achieves up to 37.4% relative character error rate reduction over verbatim baselines, offering a lighter-weight alternative suitable for deployment on consumer devices.
FormalASR addresses a practical gap in automatic speech recognition: most ASR systems are optimized to reproduce speech verbatim, including filler words and informal structures that are poorly suited for writing-oriented applications such as meeting transcription or document drafting. The conventional workaround—a two-stage pipeline pairing an ASR model with a large language model for post-editing—introduces additional latency, memory costs, and deployment complexity. To train their models, the researchers constructed two large-scale datasets, WenetSpeech-Formal and Speechio-Formal, by rewriting existing spoken Chinese corpora into formal text using LLM-based rewriting and quality filtering. They then fine-tuned Qwen3-ASR at both 0.6B and 1.7B parameter scales using supervised fine-tuning. Experiments on both datasets show improvements not only in character error rate (up to 37.4% relative reduction) but also in ROUGE-L and BERTScore metrics, indicating better semantic and textual quality. Because the formalization step is integrated into the ASR model itself, no additional language model is needed at inference time, reducing resource requirements for on-device use. The work was submitted to arXiv in May 2026 and has not yet undergone formal peer review.
What's missing
The study does not report results on spontaneous speech from domains outside its two constructed datasets, leaving generalization to other Chinese speech styles or accents unclear. As a preprint, the work has not yet been peer-reviewed.
What different sources said
- arXiv cs.AICenter
FormalASR: End-to-End Spoken Chinese to Formal Text
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.