Voice AI API for agencies building phone agents
A client says yes and wants their phone agent live soon. Each build means wiring speech recognition, a language model, a voice and a carrier together, running a server for it, and then holding bills you have to pass on. For agencies and freelancers who build phone agents for businesses on a setup fee and a retainer. Each client gets their own Wixzel Voice account, in their name, and calls go over their own carrier account. One API key per client covers every voice engine, billed per second, with Indian languages on sarvam ($0.0617 per minute) and gemini-live ($0.0491).
Last updated

How do you build phone agents for your clients with Wixzel Voice?
- 01One account per client
Open a Wixzel Voice account in each client's name. Its credit, numbers, calls, limits and API keys are its own. There are no team logins or sub-accounts yet, so you sign in to each account or keep an API key for each.
- 02POST /v1/agents/generate
Make the first draft of the agent from the client's website: its name, prompt, opening line, knowledge base and mascot. Change any of it with PATCH /v1/agents/{id}.
- 03POST /v1/sip-trunks
Connect the client's carrier account (Twilio, Telnyx, Plivo, Vonage, Bandwidth, Exotel, Vobiz or any SIP provider) so the agent answers the number they already have, or buy a new number on it with POST /v1/phone-numbers/purchase. The carrier bills the client directly.
- 04POST /v1/mcp-servers, PATCH /v1/webhook
Connect the client's tools: apps like Cal.com, Calendly and Pipedrive that the agent uses during calls, and a signed webhook to their CRM or to Zapier, Make or n8n after each call.
- 05GET /v1/usage/summary
The client's spend for any period, itemised in GET /v1/usage/events by speech, model, voice and orchestration fee. This is the cost you set the retainer against.
Which voice engine fits a client's phone agent?
Use the sarvam engine on Wixzel Voice, at $0.0617 per connected minute. Many clients' callers speak Hindi, Tamil, Malayalam or another Indian language, and sarvam is built for 11 Indian languages end to end. Where cost matters most, gemini-live is $0.0491 per minute and speaks 97 languages, Hindi, Tamil and Malayalam among them. Each client's agent can use a different engine.
What does a client's phone agent cost on Wixzel Voice?
Wixzel Voice engine time for 2-minute calls on the sarvam engine, at typical speech rates, billed per second of actual usage. Telephony is your own trunk at your carrier’s rate and is not included, and shorter calls cost proportionally less.
| Calls per month | Connected minutes | Engine cost per month |
|---|---|---|
| 300 | 600 | $37.02 |
| 1,000 | 2,000 | $123 |
| 3,000 | 6,000 | $370 |
Prepaid: add credit from $5 and the calls spend it. Estimate your own mix.
Which guides show how to build phone agents for your clients on Wixzel Voice?
Frequently asked questions
Can an agency build phone agents for its clients on Wixzel Voice?
How do I run several clients?
Whose carrier account do the calls use?
What does a client's agent cost to run?
Can I show a client a working agent before they sign?
Who pays for the credit?
Can the client see their own calls?
What else do agencies and developers build on Wixzel Voice?
Build an agent that answers the phone, checks availability and books an appointment, over your own SIP trunk, with the booking written back through the API.
Upload leads, start a campaign, and have an agent call each one over your own SIP trunk and ask your qualifying questions. Every call is a record with its status and transcript, linked to the lead.
Answer a support number with an agent that reads from a knowledge base you upload, handles the common questions, and leaves a transcript of every call.
Run phone agents in Indian languages on models built for them, with the sarvam engine covering speech-to-text, the language model and text-to-speech end to end.
Put an agent in a browser page, a Flutter app or a kiosk with no SIP trunk and no phone number: your server mints a one-minute session, the app streams audio over a WebSocket, and the session is billed like a call.
Put a talking AI assistant on your website with one script tag: an agent made from your site, with an animated mascot visitors call from the browser, caller details, who-can-call settings and daily caps.
Connect Claude Code, Codex, Cursor or any MCP client to your account, and let an AI agent connect a carrier, build a voice agent, place calls, run campaigns and read transcripts, with every tool that spends money asking first.
Run AI phone calls for a few cents a minute: per-second billing from prepaid credit, no subscription or minimum, your own SIP trunk with no telephony markup, and every charge itemised.
Give a voice agent a face: an animated avatar that lip-syncs to the agent, for a website, a share link, a web app or a Flutter app, with no fee for the avatar.
Operator, prices, limits, refunds and data location, on one page.
Build your first voice agent

