Audiobook
How do you generate a voice for an audiobook from text?
An audiobook wants a chapter, not a paragraph. the free tool reads up to 8 KB per request — about twelve hundred words — and returns WAV The whole book is slices on this page, or the API in the pipe There is no character studio and no bed.
How to do it in three steps
Pick a voice and keep it for the book. Switching voice mid-chapter breaks the listener. Vitória or Gabriel for Brazilian Portuguese.
- Cut the chapter into blocks that fit in 8 KB. Keep punctuation and dialogue. Mark the block number in the file name after download.
- Generate each block, listen to the join between the end of one and the start of the next. Tweak the last sentence if the voice cuts in the middle of a period.
- Join the WAV files in the audio editor you already use. The page does not concatenate chapters and does not insert a page-turn silence.
What to expect: formats, size and time
8 KB per request, WAV, catalogue voices, daily allowance. No cloning, no character voice by tag, no bookstore DRM. Variable time.
A book of dozens of chapters is API and disk work, not tab work. The free page exists to hear the voice's tone before you queue the book.
When to use the paid API
When the audiobook is the product. The API synthesizes block by block in the pipe. /products and /pricing.
Hear one chapter today: free page. A publisher catalogue: API.
Frequently asked questions
Does the page generate the whole book in one file?
No. Each request fits in 8 KB. The whole book is slices or the API.
Can I have one voice for the narrator and another for dialogue?
You can generate two WAV files and mix them in the editor. The page does not switch voice mid-text on its own.
Is there chapter marking for the audio app?
No. Chapter and metadata belong to your packager, after the WAV files.
Does the voice tire over a long hour?
Listen to a real block of the book, not a sample. Voice choice is yours, in the page catalogue.