Last updated: 11/10/2026Last verified: 2026-11-10
Emma is a face-to-face AI persona from Nuance Labs, now in research preview: you video-chat with her in the browser in real time, and Nuance says she reacts while you are still speaking. She runs on Nuplex, which the company launch thread calls a full-duplex audiovisual foundation model; the same technology is also offered to developers as the Nuance Live API, which that thread says is being released in beta.
What is Emma (Nuance Labs)?
Emma is a face-to-face AI persona from Nuance Labs: you video-chat with her in a browser in real time, and the in-app copy says she needs at least your microphone. The app is labelled a research preview. Nuance says she runs on Nuplex, which the company's launch thread calls a full-duplex audiovisual foundation model, and that Emma reacts while you are still speaking, which is a company claim and not a result we tested. Developers build around a Nuance persona with Nuance Live, the API documented at docs.nuancelabs.ai; the launch thread says the Nuplex API is being released in beta with a few personas.
Key Features
- Browser-based video conversation: talk face-to-face with Emma in the web app; the in-app copy says she needs at least your microphone.
- Guest access: the app offers one conversation without an account, and a further guest attempt returns a 429 guest_limit error, “You've had your free conversation”.
- Account-based continuity: signing up free is advertised as unlocking longer conversations and “Emma remembers you between calls”.
- Reactions during speech: Nuance says Emma reacts while you are still speaking. This is the company's description of its full-duplex model and we have not tested it.
- Developer access: the same sign-up dialog lists “API access to build with her model”.
- Nuance Live API integration: @nuancelabs/js-sdk handles the API calls and the LiveKit room, while your server mints a short-lived session token with an API key.
- Selectable API personas: personality.persona_id picks sam (the default), sonia or mira.
- Separate personality configuration: a persona sets only how the character looks and sounds, while personality.system_prompt says who the character is.
Best For
Pricing
No public pricing page is published: /pricing returns 404 on both app.nuancelabs.ai and platform.nuancelabs.ai, and no price appears in the app. What is visible is free to try: a guest gets one conversation, and a free account is advertised as unlocking longer conversations, memory between calls and API access. API usage is metered in credits, and the docs say “Your plan sets how much time a credit covers and how many credits you get each day”, but no plan or price is named, so API access should not be read as unlimited free usage.
Pros & Cons
Pros
- A guest conversation provides a low-friction way to try the experience.
- The browser-based interface does not require a dedicated app for the described experience.
- Free-account features are advertised as longer conversations and memory between calls.
- An API and JS SDK let developers build their own real-time experience around a Nuance persona.
- Separating persona appearance and voice from personality instructions gives developers distinct configuration controls.
Cons
- Emma is in research preview and the API is in preview too: the docs say its routes and events may change before 1.0.
- Guest use is limited to one conversation; longer conversations require a free account.
- No plan, price, conversation duration or credit allowance is published anywhere on the site.
- The API integration needs your own server to mint session tokens, plus an API key from the Nuance platform.
- Latency, audiovisual quality and how well Emma reacts during speech cannot be assessed from the published material.
Alternatives
Pre-rendered avatar video and video translation from a script, rather than a live conversation you join in the browser.
Talking-avatar video rendered from text or audio, rather than a real-time face-to-face call.
Avatar-led training and explainer videos generated from a script, not an interruptible live persona.
FAQ
Details
Platform
Features
- Face-to-face video conversation in the browser: Emma needs at least your microphone, and a guest can have one conversation with no account
- A free account unlocks longer conversations, Emma remembering you between calls, and API access to build with her model
- Nuance Live API in beta: the @nuancelabs/js-sdk handles the calls and the LiveKit room, and your server mints a short-lived session token
- Three selectable personas (sam, sonia, mira) picked by persona_id; a persona sets only how it looks and sounds, while who it is comes from your own personality.system_prompt
Languages
Known limitations
- Research preview with no published plan or price. The credits page only says “Your plan sets how much time a credit covers and how many credits you get each day” and never names a plan, and /pricing is 404 on both app.nuancelabs.ai and platform.nuancelabs.ai. Guests get a single free conversation (429 code guest_limit, “You’ve had your free conversation”), and guest access can be switched off: the app also returns 403 policy_curfew and the line “Guest access isn’t open right now.”
- The terms, last updated 9 November 2026, require 18+ and residence in the United States, Canada, the European Union or the United Kingdom, and say you are not a resident of or located in Colorado, Connecticut, Illinois or Texas
- The persona speaks English only; docs say a prompt or tools with non-English characters, or that ask it to speak another language, are refused. Healthcare and education are not in the beta, and the persona refuses medical, mental health, legal, tax or financial advice
- Conversations run on shared capacity: a start can be turned away with 429 capacity_unavailable or 503 busy, and a running conversation ends early when its runtime is taken back. The docs call the API a preview whose routes and events may change before 1.0
