ArticleTapes

Turn articles into audio.

Paste a link or text. Free. No signup.

A Your next good listen
Stays on your device

Your reading list, on tape.

Turn an article into something you can listen to on a walk, over coffee, or away from the screen. Paste a link or text and choose Make tape. Links are imported automatically. Use Import first if you want to review the text before recording.

Is article-to-audio conversion free?

Yes. The default voice runs on your device, with no account, subscription, or API key. You can optionally connect your own DeepInfra or ElevenLabs account in Settings; those providers charge you directly.

How long does it take?

Your first tape downloads the voice model. After that, generation time depends on your device and the article length. Keep this tab open while it works. Completed sections are saved so you can continue an interrupted tape.

Which articles can I listen to?

Import a public article link, choose pages from a text-based PDF, or paste up to 30,000 characters. Some sites block imports or require a login; paste the text you can access instead. English works best with the on-device voices.

Can I download the audio?

Yes. Download a finished tape as a WAV audio file. Your tapes and listening progress are saved in this browser, without cloud sync. Download anything you want to keep before clearing browser data.

Where does my article go?

With the default on-device voice, narration happens in your browser. A link import sends the URL to our server to extract the article; we do not store it there. If you connect a speech provider, the article text goes directly to that provider.

A PDF, on tape.

Text-based PDFs · up to 50 MB · 30,000 characters per tape. Scanned pages need OCR first. Your PDF is never uploaded. If you choose a connected voice provider, the imported text is sent to that provider when you make a tape.

Settings

Automatic uses GPU when available and falls back to CPU if needed. First download: about 330 MB for GPU or 90 MB for CPU. When your browser reports data-saving mode, Automatic uses the smaller model.

Voice changes apply to new listens. Interrupted listens resume with their original voice and provider.

Your library stays in this browser. Clearing site data removes it. Download anything you want to keep.

Use Article Tapes on your phone

On iPhone, open this site in Safari, tap Share, then Add to Home Screen. On Android, use Install app or Add to Home screen in your browser’s menu.

Preparing offline listening…

Open the installed app online first, then make your tapes there. Keep it open while audio generates. Tapes stay on this device; they don’t sync from your computer. On iPhone, the home-screen app has a separate library from Safari.

Your phone can remove website data when storage is low. Download tapes you want to keep.

Privacy

On-device mode: speech is generated in your browser. Model files are downloaded from Hugging Face and cached when your browser allows it. Your article text is not sent for narration.

Connected providers: your text and API key go directly to DeepInfra or ElevenLabs. Their pricing and data policies apply. You pay them for generation.

Article import: the link is sent to our server to fetch and extract the page. Login-only sites may need pasted text. We do not store imported articles on the server.

Your library: text, audio, and progress live in browser storage, with no account or cloud sync. Download audio and text before clearing browser data. Hosting providers may retain standard request logs.

On-device voices use Kokoro (Apache-2.0). All narration is AI-generated. English is recommended for on-device voices.