KeyoAPI
Eleven V4 Turbo API on KeyoAPI — Fast Speech
Eleven V4 Turbo for low-latency, high-quality text-to-speech on the same Keyo API key as chat — POST /v1/audio/speech returns audio fast.
Overview
eleven-v4-turbo is a fast neural text-to-speech model on KeyoAPI. Indicative rates about $40 per 1M characters, billed per input character — confirm live on Model Square. Call POST /v1/audio/speech with model=eleven-v4-turbo, an input string, and a built-in voice; audio comes back synchronously in the response. It is tuned for real-time interactions and large-scale generation.
Quick start
Base URL: https://www.keyoapi.xyz/v1
Endpoint: POST /v1/audio/speech
Model: model=eleven-v4-turbo
curl https://www.keyoapi.xyz/v1/audio/speech \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"eleven-v4-turbo","input":"Hello from Eleven V4 Turbo.","voice":"george","response_format":"mp3"}' \
--output speech.mp3Auth uses the same Bearer API key as chat. Full notes: Keyo docs.
FAQ
Sync or async?
Sync: POST /v1/audio/speech returns audio in the same response. Async TTS on Keyo remains Qwen3-TTS / CosyVoice3.
How is pricing metered?
eleven-v4-turbo bills per input character. Indicative ~$40 per 1M characters; confirm live rates on Model Square.
Which voices are available?
Built-in voices include george, sarah, adam, alice, bella, bill, brian, callum, charlie, chris, daniel, eric, harry, jessica, laura, liam, lily, matilda, river, roger, and will.