read text aloud online
Sometimes it is more convenient to listen to a text than to read it, during a commute, while doing something else, or just to rest your eyes. This page opens the tool with a longer sample text, showing how it handles a full paragraph.
Text of that length is exactly the scenario where competing tools tend to hang or stop midway.
Text pasted from an email often comes with odd line breaks, the tool normalizes that automatically before reading.
Because they are technologies from different eras. Voices badged "Basic" are concatenative or formant synthesizers your operating system has shipped since the 2000s: they assemble speech by gluing recorded fragments, which is where the metallic sound comes from. Voices badged "Natural" are already neural models from the system itself (Microsoft's "Natural" voices, Google's on Android and Chrome). Those badged "Neural" and "Premium" are models this tool downloads and runs in your browser, and they are the best sounding. If the current voice is ugly, the fix is to filter by "Neural" and pick one in your language.
There is no artificial limit imposed by the tool itself, the automatic chunking is exactly what allows reading text of any length without hanging. Extremely long text (a whole book, for instance) simply produces a longer queue of speech chunks, which is expected.
You can pause and resume within the same page session, but the position is not saved if you close the tab, since the reading progress itself is not persisted (only the text you typed stays saved, not the reading position). The previous/next sentence buttons help you jump back to a specific point quickly after reopening.
No voices
No voice matches these filters.
The pitch control exists only for system voices. In neural voices the timbre is part of the model, and speed is applied inside synthesis, so the voice never turns squeaky.
The system voice never hands its audio to the page, so there is nothing to export. Pick a Neural or Premium voice to produce MP3, WAV and subtitles.
All processing happens in your browser: the text, the synthesized audio, the subtitles and the pronunciation dictionary are never sent to any server. Neural voice models are downloaded from public Hugging Face repositories and kept in browser storage (OPFS), on this device only.