KeyoAPI
IndexTTS-2 API — Text to Speech and Voice Cloning
IndexTTS-2 is available on KeyoAPI as a hosted text-to-speech API: one REST call turns text into audio, with no model deployment and no separate vendor account.
Overview
IndexTTS-2 is available on KeyoAPI as a hosted text-to-speech API: one REST call turns text into audio, with no model deployment and no separate vendor account. If you need an IndexTTS-2 endpoint you can call from code today, this is the shortest path. It runs on the synchronous speech endpoint, so a request returns audio in one round trip. That makes it the simpler of KeyoAPI’s TTS paths. For long narration or batch queues, CosyVoice3 and Qwen3-TTS (async) are usually a better fit. About $1.37 per 10,000 characters—map script length to cost in your CMS. Confirm live rates on /pricing/IndexTTS-2 before locking a quote. Product voiceovers, IVR prompts, agent spoken replies, and digital-human pipelines (pair with InfiniteTalk or Duix-Avatar). Clone-friendly control of timbre and tone via audio or text prompts where the API schema allows. POST /v1/audio/speech with model=IndexTTS-2. Docs: /brand/keyo-docs.html. Try/buy: /pricing/IndexTTS-2. You do not need to deploy weights: authenticate, send your text, and write the returned audio bytes to disk or a CDN. Keep TTS next to Whisper on the same key for full duplex voice agents. Normalize numbers and abbreviations before synthesis. Cache identical scripts by content hash. Fail closed on celebrity likeness without rights. Measure listener completion, not only HTTP 200. Try it from /pricing/IndexTTS-2. There is no separate upload playground.
Quick start
Base URL: https://www.keyoapi.xyz/v1
Endpoint: POST /v1/audio/speech
Model: model=IndexTTS-2
curl https://www.keyoapi.xyz/v1/audio/speech \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"IndexTTS-2","input":"Hello from KeyoAPI TTS."}' \
--output out.mp3Auth uses the same Bearer API key as chat. Full notes: Keyo docs.
FAQ
Is there a hosted IndexTTS-2 API?
Yes. KeyoAPI serves IndexTTS-2 over POST /v1/audio/speech. You call it with an API key and pay per use — no self-hosting.
Sync or async?
IndexTTS-2 uses the synchronous speech endpoint: one request returns audio. For long-form or batch jobs, prefer CosyVoice3 or Qwen3-TTS on the async path.
How do I try it?
Create a key at /sign-up, then open /pricing/IndexTTS-2 for interactive try/buy. Free LLM IDs (separate from TTS) are listed at /free-models.
Price?
About $1.37 per 10,000 characters — confirm live rates on /pricing/IndexTTS-2.