Voice overview
Generate speech, transcribe audio, and build realtime voice agents with Inworld Voice.
Inworld Voice brings together text-to-speech, speech-to-text and the Realtime API. Choose the API for the audio task you want to build.
Text-to-Speech (TTS)
Turn text into natural speech for your app, agent or character.
Speech-to-Text (STT)
Transcribe recordings or live audio.
Realtime API
Build a voice agent that listens and talks in real time.
Make your first request
- TTS quickstart: generate speech from text.
- STT quickstart: transcribe audio.
- Realtime quickstart: start a live voice conversation.
Connect voice and inference
Use Inworld Inference for text, image, video and decision models. To generate a text answer and speak it in one API call, see Voice responses (LLM + TTS).
For credentials, limits and usage, see Go to production.