French Text to Speech: Free, In-Browser, Nothing Uploaded

French gets both neural tiers, but only one voice in each — which makes the choice between them unusually clear-cut. Paste French text, press play, and the audio is generated on your own device — no account, no upload, no character limit. This page is the honest version: which engines really speak French (Français), which ones silently do not, and what each one costs you.

Which engines actually speak French

Quick TTS ships four TTS engines and lets you switch between them from one dropdown — but not all four cover every language, and the gaps are not the ones you would guess from the marketing.

French is a two-engine language here, and the two engines are not competing for the same job:

• Piper ships one French voice, fr_FR-siwis-medium — metropolitan French, at the medium quality tier. It runs on WebAssembly, so it needs no GPU.
• Supertonic HD speaks French at 44.1kHz, as one of this build's fifteen mapped languages. It is the quality ceiling and it needs WebGPU.
• Kokoro-82M is English-only and is hidden on French — it is not an option whatever your hardware.

There is no French equivalent of the Spanish situation, where you pick between a peninsular and a Latin American voice. There is no Canadian French Piper voice in the upstream catalogue, so Quebec French is read with a metropolitan accent, and that is worth knowing before you judge the output rather than after.

The honest limits

French has no smaller Piper tier to fall back to. Only five languages in this build ship a working step-down model — English, Spanish, Polish, Dutch and Arabic — and French is not one of them. If a device cannot allocate fr_FR-siwis-medium, which is the usual failure on low-RAM machines, French drops to the Web Speech API rather than to a lighter neural voice. The smaller upstream French models are not shipped because they fail at inference in this runtime; offering one and having it break would be worse than not offering it.

Supertonic HD's cost is a one-time ~380MB model download, cached afterwards so return visits fetch nothing, plus a hard WebGPU requirement with no WebAssembly fallback tier. On a desktop with a working GPU that is the best French available here by a clear margin. On a phone or in Safari, the realistic French options are Piper and your operating system's built-in voice — and macOS, iOS and French Windows installs all ship a decent one.

French is also the language where the accent-and-liaison problem shows up fastest: run the same paragraph through the system voice and through Supertonic and the difference is mostly in liaison and elision, not in raw audio quality — which is the thing a sample-rate number on a spec sheet will never tell you.

Nothing you paste is uploaded

This is the part that separates in-browser TTS from the cloud tools that look identical in a browser tab. A cloud tool sends your French text to a server, synthesizes it there, and sends audio back — the page is a remote control. Quick TTS ships the engine to you instead and runs it on your CPU or GPU, so your text is never received by anyone. The test is simple: once the page and its model have loaded, the neural engines keep working with your network disconnected.

That matters most for the documents people actually want read aloud — a contract, a medical letter, coursework, an unpublished draft. It also means there is no "we phonemize non-English text on our server" asterisk, which is not true of every tool advertising local synthesis.

Reading French documents, not just pasted text

PDF, DOCX, EPUB, ODT, RTF, HTML, TXT and Markdown are all parsed in the browser, so a whole French EPUB or a long PDF opens and plays without a byte leaving your machine. A scanned PDF — a photo of text with no text layer — is handled by running OCR locally in WebAssembly rather than uploading the scan, which is precisely the case where local processing is worth the most.

Speed, volume and voice can all be changed mid-read and the current passage replays under the new settings without starting over. Switching the engine itself stops playback, so you press play once to start the new one — there is no mid-read handover of the remaining text between engines.

Frequently asked questions

Is French text-to-speech free?

Yes. Quick TTS reads French with no account, no character limit and no watermark. Synthesis runs in your browser on your own device, so there is no per-character server cost to meter and nothing to gate behind a sign-up.

Is my French text uploaded anywhere?

No. The speech engine ships to your browser and runs locally, so the text you paste is turned into audio on your machine and never sent to a server. That is an architectural property, not a promise in a privacy policy — there is no server in the loop once the page has loaded.

What is the best French TTS voice in the browser?

Supertonic HD is the highest-quality French voice available here — 44.1kHz, running on WebGPU. If your machine has no usable GPU, Piper is the neural option that still runs. To get French voices rather than English ones, open the app at quick-tts.com/fr/ — the neural voice list follows the page language.

Does it work on a phone?

The Web Speech API engine works on every modern phone with no download, and it is the default so the first play click always works. The neural engines are more demanding: Supertonic HD needs WebGPU and is effectively a desktop feature, and Piper runs more widely but can be slow on low-end hardware.

Can I download the French audio?

Yes, as WAV or MP3, generated on your device. Anything longer than about 20,000 characters is packaged as a ZIP of audio parts so memory stays bounded on long documents.

Try it in French — open the Français app, not the English one

This is the one piece of setup worth knowing, because it is not obvious: the neural voice list follows the language of the page you are on. On the English homepage the Piper voices are the English ones, and Supertonic reads with the English language tag — so French text pasted there will be pronounced as though it were English. Open the Français version instead and the engine picks up French voices, including fr_FR-siwis-medium, automatically.

The one exception is Supertonic's "Auto language" toggle, which switches generation to a language-agnostic mode that reads mixed-language text with the right pronunciation per language without tagging anything by hand. That works from any page — it is the right choice for a document that is part French and part English.

Written in Français

These are the French-language write-ups on this site. They are authored in Français rather than translated from the English, so they are the better read if Français is the language you would rather be reading in:

• Les meilleures voix TTS françaises 2026 : Hortense, Thomas et Edge Online comparées
• Synthèse vocale HD gratuite dans le navigateur, sans cloud

Each of this language's Piper voices also has a page of its own, covering the quality tier, what happens when your device cannot load the model, and how it compares to the HD option: fr_FR-siwis-medium.

For the architecture behind all of this, see how in-browser text-to-speech works; for a tool-by-tool comparison with the other local TTS sites, see the comparison page.