Build, monitor, and ship agents from one console
Build and monitor agents from one console, built in-house on our own stack.
The production console, as it ships.
Build the agent
Compose a voice agent in the console — voice, knowledge, and behavior — with no separate tooling.
Monitor every call
Watch live call sessions, transcripts, and analytics, and catch gaps as they happen.
Ship on your own stack
Deploy through the same guarded pipeline the whole platform runs on — a human is the only gate to production.
Improve in plain language
Refine behavior and correct the agent over time from the console.
One connected flow
One call, followed end to end in the console
The transport connects, the conversation is transcribed live, and the record and its follow-ups hand off to the console. Redaction and risk scoring are already done while that call is happening — not in a batch afterwards.
Browser call
12m 04s
Duration
Live transcript
Member #1042 · redacted
- Agent · Hello — just calling to check in with you today.
- Member · Yes, I'm doing all right.
- Agent · Take your time; I'll wait.
Personal identifiers are stripped inside the same pass that handles the turn.
Ask the care lead to confirm
Care team
Member
Member #1042Active
Recurring check-in · Next call Thursday at 10:00
Calls
Transport
Browser · WebRTC
The mobile app runs against the same pipeline
Console screens
Every capability, honestly labeled
Each capability with its real maturity — production, beta, or in development — measured against the actual repository, not a roadmap.
See the platformText to Speech
Turn text into lifelike speech
Speech to Text
Convert audio and video into precise, editable text
Voice Isolator
Remove background noise, music, and interference to leave clean, professional speech
My Voices
Voices you own — cloned, designed, and saved from the library
API Keys
Create keys to authenticate API requests
Webhooks
Receive events when generations finish, calls complete, or transcriptions are ready
Request Log
The most recent API requests with method, status, latency, and key
Sound Effects
Describe a sound in plain language and get four royalty-free samples in seconds
Build and monitor
The place you build it is the place you watch it
The screen where an agent is composed and the screen where you watch that agent take a call live in the same console.
Build the agent
Compose a voice agent in the console — voice, knowledge, and behavior — with no separate tooling.
See how it is composedMonitor every call
Watch live call sessions, transcripts, and analytics, and catch gaps as they happen.
See how monitoring works1const session = console.createSession({2 locale: "ko-KR",3 transport: "webrtc",4});56session.pipeline([7 transport,8 stt,9 korean_turn_tuning,10 llm,11 tts,12]);1314// Transcripts are redacted before the console records them.15session.on("transcript", (turn) => {16 record(turn.redacted);17});Creative
Generate speech and manage voices
Agents
Deploy and monitor voice agents
Developers
API keys, webhooks and usage
Screens
Member #1042
Recurring check-in · Browser
Member #0887
Follow-up call · Mobile
Member #1130
Recurring check-in · Browser
Member #0471
Recurring check-in · Mobile
Example screen.
Put a voice agent to work in care
Book a walkthrough of WelVoice and the operator console — voice agents for care calls on your own stack.
Guides
From rollout design to day-to-day operations, in the order teams actually work.
Read the guidesEvery console screen on this page is an example. It shows no captured media, no customer, and no real care setting, and member rows are masked identifiers. Everything that reaches production passes one human gate. IntuneLabs holds no security or privacy certification.