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

Lithuanian has zero Piper voices upstream and no Kokoro support, so the honest story here is one desktop-only neural engine plus whatever your OS already ships. Paste Lithuanian 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 Lithuanian (Lietuvių), which ones silently do not, and what each one costs you.

Which engines actually speak Lithuanian

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.

Only one of Quick TTS's three neural engines speaks Lithuanian, and the other two are worth being specific about:

• Piper ships zero Lithuanian voices. The upstream Piper voice catalogue has no Lithuanian model at all — not a low-quality one, not an experimental one, none. A Lithuanian model has been requested against the project, but nothing has shipped, so there is no lt_LT-* voice name to offer.
• Kokoro-82M is English-only and is hidden on every non-English locale this site serves, Lithuanian included.
• Supertonic HD does speak Lithuanian, natively, at 44.1kHz, as one of the 31 languages this build maps.

That leaves a two-tier reality: a desktop GPU gets a genuinely good neural voice, and everything else falls back to whatever Lithuanian voice the operating system happens to include — which, on Windows in particular, is not a safe assumption.

The honest limits

Supertonic is WebGPU-only with no WebAssembly fallback tier, and there is no Piper voice sitting underneath it the way there is for languages like Spanish or Polish. On a desktop with a working GPU you get studio-grade 44.1kHz Lithuanian after a one-time ~380MB cached download. On a phone, or a desktop without WebGPU, you get whatever your operating system ships, and nothing in between.

Be precise about that floor: a default Windows install frequently ships no classic SAPI Lithuanian voice at all, though Microsoft Edge's online Read Aloud feature adds neural Lithuanian voices (Ona and Leonas) that stream from Microsoft's servers rather than running offline. Android's Google text-to-speech engine supports Lithuanian once the language pack is installed. So the realistic picture for most Lithuanian readers on a phone or an older Windows machine is Edge's online voice or Android's system voice, not a fully offline neural engine — that only exists on desktop with WebGPU.

Lithuanian's nasal vowels — ą, ę, į, ų — and its free, phonemic pitch accent are the two things that trip up every TTS engine trained mostly on other languages. A word's meaning can hinge on which syllable carries the acute or circumflex accent, and ordinary spelling does not mark it, so an engine has to guess from context the way a human reader does. If you are proofreading Lithuanian by ear, treat an odd stress on an unfamiliar word or proper noun as a prompt to check the spelling rather than proof the engine is simply wrong.

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 Lithuanian 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 Lithuanian documents, not just pasted text

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

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

Supertonic HD is the highest-quality Lithuanian voice available here — 44.1kHz, running on WebGPU. It is also the only neural Lithuanian option: Piper ships no Lithuanian voice and Kokoro is English-only, so the alternative is your operating system's built-in voice. To get Lithuanian voices rather than English ones, open the app at quick-tts.com/lt/ — 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.

Can I download the Lithuanian 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 Lithuanian — open the Lietuvių 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 Supertonic reads with the English language tag — so Lithuanian text pasted there will be pronounced as though it were English. Open the Lietuvių version instead and the engine picks up Lithuanian 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 Lithuanian and part English.

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.