New Framework Enables AI Systems to Plan and Express Emotions in Real-Time Speech
Researchers have proposed Self-EmoQ, an emotion-planning framework that determines appropriate emotional tone before text generation to drive streaming text-to-speech synthesis in conversational AI. The system uses a plug-and-play large language model module trained via reinforcement learning, guided by Plutchik's wheel of emotions as a theoretical foundation. The work, accepted to ACL 2026 Findings, addresses a gap in current conversational AI systems that lack autonomous emotion-determination mechanisms for real-time speech output.
Self-EmoQ is a novel framework designed to give conversational AI systems the ability to autonomously determine emotional tone prior to generating text, which then drives downstream streaming text-to-speech synthesis. The system is implemented as a plug-and-play module built on pretrained large language models and trained using reinforcement learning, with emotions treated as discrete actions. A hybrid reward function combines imitation learning signals with theory-driven scoring grounded in Plutchik's wheel of emotions, a well-established psychological model of emotional structure. Experiments conducted on four benchmark datasets—DailyDialog, EmoryNLP, IEMOCAP, and MELD—show that Self-EmoQ outperforms both prompting-based and fine-tuning baselines on emotion determination accuracy and response quality. The researchers also implemented a full end-to-end streaming pipeline suitable for real-time deployment, with evaluations confirming emotional alignment, contextual coherence, and expressive fluency in the synthesized speech. The paper has been accepted to ACL 2026 Findings, and code, case studies, and demos have been made publicly available.
What's missing
Generalizability beyond English-language datasets is not discussed. Additionally, human perceptual evaluation methodology and inter-rater reliability for subjective speech quality assessments are not detailed in the abstract.
What different sources said
- arXiv cs.AICenter
Self-EmoQ: Plutchik-Guided Value-based Planning to Drive Streaming Emotional TTS
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.