Nuance Labs raises $50M Series A to build real-time audio-visual emotion foundation model
soleio · x · 2026-09-15
Nuance Labs announced a $50M Series A led by Lightspeed, with participation from Accel, SPC, NVIDIA and Define VC.
The startup is building a foundation model that understands and responds to the full spectrum of human emotion: a single full-duplex audio-visual model that sees, listens and responds in real time, mimicking the natural back-and-forth of a conversation with a friend. The founders argue most human communication happens beyond words—faces, smiles, gestures—and that this visual-emotional vocabulary is the key to a better machine interface.
Commenters called the demo startling, framing it as a step toward AI that is both safe and helpful: to follow intent, AI must understand nuance—hence the company's name.
More from Venture
- AI Accounting Meme: EBITDARSWDPMTC Excludes Model Training Costs — prateekj · 2026-09-15
- Sequence and Dell Family Office take Baldwin Group private in $7.7B deal, 18 months after seed — saranormous · 2026-09-15
- OpenAI quietly bought a smartphone camera startup for ~$300M — KateClarkTweets · 2026-09-15
- Superhuman acquires AI meeting notetaker Fathom to turn conversations into action — matt_slotnick · 2026-09-15
- Palo Alto Networks CEO: 'It's over' for analytical SaaS companies as LLMs replace them — rohanpaul_ai · 2026-09-15
- SoftBank Borrows $11.87B for OpenAI Stake as Shares Slump 13% — eyishazyer · 2026-09-15