Transcribers (Speech-to-text)

This page lists the speech-to-text (STT) providers Vapi supports, with each provider’s ID.

The transcriber converts the caller’s speech to text for your voice agent. You select one by setting transcriber.provider and transcriber.model on an assistant. See Create Assistant in the API reference.

Language coverage and automatic detection behavior can vary by model within the same provider. Review the provider page before choosing a model and language combination.

Quick reference

The provider string is the exact value you pass in transcriber.provider.

Provider	Provider string
AssemblyAI	`assembly-ai`
Azure	`azure`
Cartesia	`cartesia`
Deepgram	`deepgram`
ElevenLabs	`11labs`
Gladia	`gladia`
Google	`google`
OpenAI	`openai`
Soniox	`soniox`
Speechmatics	`speechmatics`
Talkscriber	`talkscriber`
xAI	`xai`

Create an assistant

Set transcriber.provider and transcriber.model in the API.