OpenAI
Run a speech-to-speech assistant, or use an OpenAI model in a cascading pipeline.
A voice stack.
Your choices.
Choose speech, models and telephone routes independently. Each guide explains the adapter APIFleet actually implements.
Two realtime providers. Two transcription adapters. Nine speech-synthesis adapters. Four telephone carriers. HTTP tools and webhooks connect the rest of your application.
Realtime voice · Language modelExplore OpenAI →
Gemini
Use Gemini Live for speech-to-speech, or choose Gemini adapters within a cascading assistant.
Deepgram
Combine streaming transcription with a separately selected voice and language model.
ElevenLabs
Add an ElevenLabs voice to APIFleet’s cascading voice pipeline.
Cartesia
Use a Cartesia voice while keeping prompts, tools and call routing in APIFleet.
Azure Speech
Speak with Azure voices from a cascading assistant.
Amazon Polly
Use an Amazon Polly voice as the output stage of a cascading assistant.
Soniox
Connect the implemented Soniox speech output adapter to a cascading assistant.
Piper
Use an operator-configured Piper service for speech output.
Telnyx
Route Telnyx telephone calls to an assistant and optionally use Telnyx speech synthesis.
Twilio
Connect Twilio media streams and call routing to APIFleet assistants.
Voximplant
Connect Voximplant call handling to the APIFleet voice runtime.
VoIPcloud
Use the APIFleet VoIPcloud carrier integration for your telephone route.
HTTP tools
Let an assistant call your application for availability, records or other explicit actions.
Webhooks
Send APIFleet events to your application for downstream workflows.
Provider access, regional availability and limits depend on your account. Estimate the cost of your combination →
Bring your next conversation.
Explore the API, or sign in to your existing workspace. New workspace access is currently limited; contact us to discuss your project.