Text to Speech
Turn text into natural-sounding speech with 40+ AI voices and download it as MP3.
Frequently Asked Questions
How do I convert text to speech?
Press play on a few voices to hear samples, select the one you like, paste your text, and click Generate speech. The audio plays right away and you can download it as an MP3.
What is the difference between cloud voices and in-browser voices?
Cloud voices run on Deepgram Aura 2 through our API: English and Spanish, up to 500 characters per request, with a shared free daily allowance. In-browser voices run the open-source Kokoro model on your own device: English only, up to 5,000 characters, no daily limit, and your text never leaves your computer. The first use downloads a ~90 MB model, which can be slow on older computers.
Can I use the generated audio commercially?
Audio from the in-browser Kokoro voices is produced by an Apache 2.0 licensed model. Cloud voices are generated by Deepgram Aura 2 on Cloudflare Workers AI; check their terms for your use case.
Why did I get a daily quota message?
Cloud voices share a free daily allowance across all visitors. When it runs out, it resets at 00:00 UTC. In the meantime, switch to in-browser voices, which have no limit.
Is my text stored?
In-browser voices never send your text anywhere. Cloud voice text is sent to the ToolDock API and processed by Cloudflare Workers AI through Cloudflare AI Gateway, which keeps request logs and caches responses for up to 7 days. ToolDock does not store your text in its own database.
Can I use this from code?
Yes. POST /api/v1/audio/speech accepts the OpenAI speech request format and returns MP3, so the OpenAI Python and JavaScript SDKs work by setting the base URL to the ToolDock API. GET /api/v1/audio/voices lists every model and voice.