Live
WebRTC (browser)
Transports
Runs a full realtime voice session in the browser — no install, and the same pipeline, processors, and flags the mobile app runs against.
- Browser voice
- No install
- Realtime
Voice engine
Integrations
What the stack actually connects to
This is not a partner list. These are the transports, model roles, and platforms the IntuneLabs stack actually connects to, and each one carries the state that connection is really in — live today, or built and key-gated.
Why it reads this way
Built and shipped are written down separately
Browser voice and mobile voice are the transports running in production today. Telephony is built but key-gated, and it is not carrying production phone calls. This page does not blur that difference.
Transports
The paths a call actually travels. Browser voice over WebRTC and native mobile voice are live today; the telephony transport is built and key-gated.
- WebRTC (browser)
- Mobile transport
- Telephony (key-gated)
Model roles
Best-of-breed vendor models, selected per role for speech-to-text, language, and text-to-speech. We don't claim to own the models, and we don't depend on any single one.
- Selected per role
- Sticky TTS failover
- Not trained by us
Platforms
The surfaces the same pipeline reaches. iOS, Android, web, and the care-operations backend each read from one pipeline rather than a per-surface fork.
- iOS
- Android
- Web
- Care backend
State labels
An entry's state is only one of live, key-gated, or planned. It is never a certification level and never a commercial relationship — no name here implies partnership or endorsement.
- Live
- Built, key-gated
- Planned
What are you looking for
Four ways into the list
Pick one and the list below narrows to it — or search across all of them.
Transports
The paths a call actually travels, with what is live today marked separately from what is built and key-gated.
Model roles
The vendor models orchestrated per role across speech-to-text, language, and text-to-speech.
Platforms
The iOS, Android, web, and care-operations surfaces the same pipeline reaches.
Product surfaces
The operator console, console API, call records, and feature flags that handle what a call produced.
Directory
Search the integration directory
Filter by kind, surface, and state. Everything here runs locally in the page — no request leaves your browser.
16 entries
- Live
Realtime media transport
Transports
The realtime media layer that carries audio for a browser session. The pipeline runs on top of it and is not pinned to any one provider.
- Media
- Browser voice
Voice engine - Built, key-gated
Telephony (key-gated)
Transports
The telephony transport is built but key-gated. It is not handling production phone calls today, and it is listed here as built, not as shipped.
- PSTN
- Twilio
- Key-gated
Voice engine - Live
Mobile voice
Transports
The native mobile app runs the same realtime voice stack, so a change to turn-taking or failover lands on both surfaces.
- Mobile voice
- Realtime
Web and mobile apps - Live
Speech-to-text (role)
Model roles
The role that turns what was said into text. Vendor models are selected per role and orchestrated; we do not train them.
- Vendor model
- Selected per role
Voice engine - Live
Language model (role)
Model roles
The role that carries the conversation. Our work is the orchestration and the layer around it; we don't claim to own the model.
- Vendor model
- Orchestration
Voice engine - Live
Text-to-speech (role)
Model roles
The role that turns text into speech. Korean text normalization runs before synthesis, so spoken output reads the way a Korean speaker would actually say it.
- Vendor model
- Korean normalization
Voice engine - Live
Sticky TTS failover
Model roles
When a synthesis vendor stops responding mid-call, the pipeline switches to another one and stays there for the rest of that call instead of flapping.
- Failover
- Call continuity
Voice engine - Live
iOS
Platforms
The native iOS app. Voice is live on it today, and it shares the pipeline.
- Mobile
- Native
Web and mobile apps - Live
Android
Platforms
The native Android app. Voice is live on it today, and it shares the pipeline.
- Mobile
- Native
Web and mobile apps - Live
Web
Platforms
The browser surface. It starts with no install and runs against the same processors and the same flags.
- Browser
- No install
Web and mobile apps - Live
Care operations backend
Platforms
Domain modules spanning care and consumer verticals, so schedules, records, and notifications move through one data flow.
- Domain modules
- Care operations
Backend - Live
Operator console
Product surfaces
Calls, transcripts, status, and follow-up actions in one screen.
- Console
- Call records
Operator console - Live
Console API and keys
Product surfaces
Per-deployment keys for the console surfaces, available in a Production or Enterprise engagement.
- API
- Keys
Operator console - Live
Feature flags
Product surfaces
Individual capabilities can be turned down one at a time, so a failing component costs a feature rather than the whole call.
- Flags
- Staged rollout
Backend - Live
Call records
Product surfaces
The redacted transcripts, risk signals, and per-call timings produced by the pipeline collect here.
- Redacted transcripts
- Per-call timings
Operator console
One level deeper
Browser and mobile are live; telephony is built
That is the distinction this page exists to keep. What follows is what it actually means — not a customer story.
What is live today
Browser voice over WebRTC and native mobile voice are the transports running in production today. Both surfaces run against the same pipeline, processors, and flags.
What is built and key-gated
The telephony transport modules are built but key-gated. They are not running as production phone calls, and telephony is integrated per deployment rather than switched on out of the box.
What we do not own
Speech and language models come from outside vendors, selected per role. IntuneLabs does not claim to have trained them, and does not depend on any single one.
IntuneLabs does not train its own speech or language models. It orchestrates best-of-breed vendor models and builds the layer around them — Korean tuning, turn-taking, failover, guardrails, and per-call instrumentation. That layer is the product.
- The transports running in production today
- Browser and mobileThe transports running in production today
- Telephony, built but not carrying production calls
- Key-gatedTelephony, built but not carrying production calls
- Vendor models orchestrated rather than trained
- Selected per roleVendor models orchestrated rather than trained
Tell us what you need to connect to
Describe the workflow and the integration surface you have in mind, and we will start from what is live today.
See the platform
How the pipeline, processors, flags, and transports fit together.
See the platformStart with the docs
How the integration surfaces and the console API actually behave.
Read the docsDeployments for care teams
Check-in calls, inline processing, and the follow-up work in one flow.
Learn more
Nothing on this page implies partnership, endorsement, or certification. Each name identifies a transport, model role, or platform the IntuneLabs stack actually connects to, and the state label records only how far along that connection is.
Every mark on this page is original SVG geometry. No third-party logo or brand asset is loaded anywhere on this surface, and names are set as text.