How it works

How a book becomes speech on your device

Open a file in your browser. The app keeps its structure, speaks it a sentence at a time, and follows along with a highlight. No server is involved.

From file to voice

1

Open

Choose an EPUB, PDF, TXT, HTML file or a public web article. The app keeps the chapters, headings and reading order.

2

Listen

The text is spoken a sentence at a time, so playback can start before the whole chapter is ready.

3

Follow

The spoken sentence is highlighted. Click a sentence to continue from there.

4

Return

Your library and reading position are kept in your browser, so you can pick up where you stopped.

Format handling

FormatWhat worksLimit
EPUBChapters and reading orderDRM-free only
PDFSelectable textA scan needs OCR first
TXTPlain text with simple chapter detectionMarkdown is not parsed
HTMLHeadings, lists and tablesScripts not executed
URLPublic articlesPaywalls and logins come through empty

The voice

The voice runs in your browser. There is no cloud service and no per-minute fee. It is English-only today.

Speed

Speech is generated as it plays, so a chapter begins with its first sentence rather than after the whole book. Measured in testing:

  • A desktop computer generated speech more than twice as fast as it plays.
  • A budget Android phone stayed faster than real time.

Results vary with the device, the browser and the text.

Privacy boundaries

  • Your files and the text read from them stay in the browser
  • No upload endpoint and no account
  • The speech runtime and voice download once, then work offline
  • A URL import fetches that URL directly
  • The host sees normal request metadata

Local voice vs cloud read-aloud · What works offline · Compare approaches.

What this does not prove

It does not prove that every device performs the same, or that the voice suits every text. Performance depends on your device and browser. The app is English-only today.

Limits

  • No OCR, DRM, Markdown, MOBI, AZW3 or DOCX support.
  • No audiobook or read-along export in the current UI.
  • Generated audio is not saved as a library.
  • The source is not publicly distributed.

Questions

Short answers to the things people ask before opening a file.

Does it need a powerful computer?

No. It runs in a current browser on ordinary laptops and phones. Performance varies by device.

Do I need to install anything?

No. It runs in the browser and can be added to your home screen.

Does it work offline?

Once the app and its voice are cached, yes. The first visit needs internet.

Does it use the cloud?

The voice runs on your device. There is no cloud voice service. Loading the app and its voice, and any URL you import, use the network.

Is the source public?

No. The repository is private. Third-party components have their own licenses.

Open a book and watch it work

The proof is simple: no upload, and your files stay on your device.

Open a book