Getting Started

Getting started

This page takes you from first launch to hearing a document read aloud. It assumes you've followed Installation.

1. Open the app

Run npm run dev. Your browser opens the Library. The sidebar (a bar along the bottom on phones) has four pages: Library, Voices, Voice Studio and Settings.

At the bottom of the sidebar, a status line says Voices ready once the Kokoro voice model is installed. If you see the banner Install the voice model to start listening instead, use its Install Kokoro button (it shows the download size) and wait for it to finish.

2. Add a document

Choose whichever suits you:

• Upload document and pick a PDF, DOCX, EPUB, Markdown or TXT file.
• Drag a file anywhere onto the Library page.
• Paste text, give it an optional title, and choose Add to library.
• In an empty library, Try a sample adds an excerpt from "A Scandal in Bohemia" by Arthur Conan Doyle.

The text is extracted on your computer; large PDFs and books can take a few seconds. The reader opens when it's done. See Reading Documents for formats and limits.

3. Choose a voice

The default reading voice is Heart, a US English voice. To change it:

• click the voice button in the player bar at the bottom of the screen and pick from the list, or
• open Voices, preview a few, and choose Use for reading.

Either way, your choice becomes the default for every document. Pick a voice in the same language as your document. See Voices.

4. Press Play

Press Play at the top of the reader, the play button in the player bar, or the Space key. The first sentence usually starts about a second later. Sentences are generated just ahead of you and cached, so even a whole book starts straight away.

To start from somewhere else, click any sentence in the text.

5. Follow along

The sentence being spoken is highlighted, and the page scrolls to keep it in view. If you scroll away yourself, the reader waits a few seconds before bringing the sentence back. You can turn this off with Follow along while listening in Settings → Playback.

The player bar stays at the bottom of every page:

Control	What it does
Previous and next sentence	Step back or forward one sentence
Previous and next section	Jump between chapters, sections or pages (on wider screens)
Position in document (slider)	Move anywhere in the document; shows elapsed and remaining time
Voice	Change the reading voice
Speed (for example 1×)	Playback speed, from 0.75× to 2×
Export audio	Save the narration as a file; see Exporting Audio
Keyboard shortcuts	Show the shortcut list (on wide screens)

Where your browser supports it, the lock screen and other system media controls can play, pause and skip (the track buttons move one sentence back or forward). See also Keyboard Shortcuts.

6. Come back later

Your position is saved automatically as you listen. Next time:

• the Library shows a Continue listening card with a Resume button for the most recent document you haven't finished;
• each document in the library shows how much you've heard;
• opening a document returns you to where you stopped, and the reader's button says Resume. Playback doesn't start until you press it.

Next steps

• Reading Documents: chapters, scanned PDFs, navigation, speed and pace.
• Voices: compare voices on your own passage.
• Voice Studio: create a voice from your own recordings.
• Exporting Audio: save narration as MP3 or WAV.