Text to speech for audiobooks and long reads
Hear a novel draft, turn a paper into commute listening, or narrate a newsletter. Because synthesis is local, unpublished work stays on your machine.
One-time download, stored in your browser and reused offline. Manage or remove it any time on the Models page.
Preloaded with Kokoro-82M — The quality benchmark for in-browser speech.
Book-length without a bill
Cloud APIs can run hundreds of dollars on a novel. Here you generate chapter by chapter on your own hardware.
Manuscripts stay private
Nothing is uploaded or used for training. The only copy is the one on your disk.
Voices that hold up for hours
Kokoro narrators like Fable, George, and Emma are solid picks for long listening.
Audiobooks & long reads — questions
Can I sell audiobooks made with these?
Check the model and voice terms, then confirm the current synthetic-narration policy of every retailer or distributor you use.
How do I handle a whole book?
Generate per chapter (up to 5,000 characters per run), download each WAV, and stitch in any audio editor. Collections help keep chapters organized.