[ About ]
A free tool, because it should exist.
No account, no paywall, no watermark, no trial. There is no revenue model and none is planned — this exists because it's useful.
What it does
Paste text, pick a voice, hear it back immediately, and download the MP3 alongside timed subtitles — SubRip (.srt) and WebVTT (.vtt) — cut from the same synthesis. That's the whole tool. Subtitle export is the lead feature, not an afterthought: the synthesis service reports exactly when each word was spoken, and almost nothing else in this category exposes that data to the person using it.
What it won't promise
Neural voices are licensed by whoever trained them, and this tool isn't in a position to grant rights it doesn't hold. If your use is commercial, check the licensing terms that apply to the voice you picked. That's a real limitation, stated plainly rather than worked around.
The interface is English only, with no plans to translate it. The output is a different matter — the voice catalogue covers roughly 150 languages and locales, fetched live rather than hard-coded, so what's listed on this site is whatever the catalogue currently offers.
How it's built
An Astro site on Cloudflare, with a Cloudflare Worker handling synthesis. There's no account system and no database behind the tool itself — nothing you submit is tied to you. (The blog you're reading this next to does use a small database for its own view counts and search, which is a separate, scoped exception — see the FAQ if you're curious what is and isn't stored.)
Notes on the build
Written up as we went, at /blog — how word-level timing actually works, what separates SRT from VTT, and what the formats can and can't carry. Everything there is something the code actually does.