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

Turkish gets both neural tiers, and no safety net under either — which is worth knowing before you judge it on an old laptop. Paste Turkish 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 Turkish (Türkçe), which ones silently do not, and what each one costs you.

Which engines actually speak Turkish

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.

Turkish is covered by both neural engines here, which is more than most in-browser TTS tools manage:

• Piper's tr_TR-fettah-medium runs on WebAssembly with no GPU required, after a one-time model download.
• Supertonic HD speaks Turkish at 44.1kHz on a WebGPU machine.
• Kokoro-82M is English-only and is hidden on Turkish.

Turkish is a fair stress test of whether a model was actually trained on the language. It is agglutinative: a single word carries suffix after suffix, so evlerimizden ("from our houses") is one token where English needs four, and real text is full of words far longer than that. An engine that has only seen Turkish in passing tends to break those words at the wrong place or flatten the stress, which in Turkish normally lands on the final syllable and shifts as suffixes are added. The other giveaway is the vowel set — the dotless ı against i, and ö/ü. A model approximating Turkish from a neighbouring language collapses those pairs, and it is immediately audible.

The honest limits

The honest limitation for Turkish is that neither neural tier has anything underneath it. Turkish is not one of the five languages with a working smaller Piper model — the upstream catalogue ships no sub-medium Turkish voice at all — so if a device cannot allocate tr_TR-fettah-medium, which is the usual failure on low-RAM machines, there is no step-down model to fall back to and Turkish drops to the system voice. Supertonic HD has no WebAssembly tier by design, so it is WebGPU or nothing, plus a one-time ~380MB cached download.

So the realistic shape for Turkish is two neural options that both work well on capable hardware and both disappear at once on weak hardware, with the Web Speech API underneath. That tier is decent for Turkish on Android and on macOS/iOS; a default Windows install often has no Turkish voice at all, and there the browser simply has nothing to speak with. On those machines Piper is not a compromise — it is the only thing that works without a GPU.

One delivery note specific to this voice: tr_TR-fettah-medium exists only in the Piper voice fork this build uses, and is not present on the mainland-China mirror that the other voices fall back to. Users behind a throttled connection to the primary host therefore have a harder time fetching Turkish than, say, Russian or Polish — the download falls through to the primary host rather than failing outright, but it is the slower path.

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 Turkish 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 Turkish documents, not just pasted text

PDF, DOCX, EPUB, ODT, RTF, HTML, TXT and Markdown are all parsed in the browser, so a whole Turkish 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 Turkish text-to-speech free?

Yes. Quick TTS reads Turkish 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 Turkish 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 Turkish TTS voice in the browser?

Supertonic HD is the highest-quality Turkish 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 Turkish voices rather than English ones, open the app at quick-tts.com/tr/ — 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 Turkish 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 Turkish — open the Türkçe 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 Turkish text pasted there will be pronounced as though it were English. Open the Türkçe version instead and the engine picks up Turkish voices, including tr_TR-fettah-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 Turkish and part English.

Written in Türkçe

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

• Tarayıcıda ücretsiz HD metinden sese — bulut yok, yükleme yok

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: tr_TR-fettah-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.