Home / Use case

Voice, speech and audio

“I need text-to-speech, transcription, or a voice agent.”

Speech in and speech out. Some run fully local, some are cloud APIs.

Tools for this

#1 · Our pick

Best for: Broadest multimodal menu

GPT models plus image, speech and realtime voice APIs under one key.

Best for: Speed

Very fast inference for open-weight models. Pick it when latency matters more than the model.

Best for: Lifelike voices

Natural-sounding text-to-speech, voice cloning and voice agents by API.

Best for: Free local transcription

Open-source speech-to-text that runs on your own machine. Free and accurate.