Loading page
Loading page
Communication · Twilio alternative
An open-source telephony stack for building AI voice agents and voice apps on real phone lines.
Fonoster is an open-source platform for building and running voice applications. The project calls itself "the AI that answers the phone". It includes a SIP server, a media engine, Asterisk, an API server and a dashboard, so you can take calls on real phone lines and control them from code.
There are two ways to build. Autopilot lets you describe a voice agent in YAML: pick the speech-to-text, the voice and the model, write the prompt, and deploy it. Voice Applications give full control: Fonoster opens a WebSocket session to your code for each call, and you say, gather, play, transfer or hang up.
Fonoster suits developers building phone bots, call routing and AI agents who want an open alternative to closed voice APIs. It needs telephony know-how, such as SIP and RTP ports, so it is a tool for technical teams.
Hosted voice APIs charge per minute on top of carrier costs. Running Fonoster yourself keeps call logic, recordings and data in your own service.
The trade-off is weight and stability. This template runs about ten containers, including Postgres, InfluxDB and NATS. Voice traffic needs open UDP ports and a public IP. The project also warns of breaking changes during current upgrades. You handle updates, backups and the carrier setup.
Containers: fonoster/dashboard:0.15.15, fonoster/apiserver:0.15.15, fonoster/autopilot:0.15.15, fonoster/routr-one:2.13.13, fonoster/rtpengine:0.3.17, fonoster/asterisk:20, postgres:16.10-alpine3.22, influxdb:2.7, nats:2.11.8, envoyproxy/envoy:v1.35.0
Step 1
Choose the Starter plan. It costs $20 a month and runs up to 3 apps.
Step 2
Open App templates in the dashboard, choose Fonoster, and click Deploy. We set up the containers, secrets and domain.
Step 3
Wait a few minutes for the first deploy, then open your new URL and finish the Fonoster setup screen.