ELEVENLABS
ElevenLabs Instant Voice Cloning
The current production option. A short sample creates a voice that can be previewed and selected for premium document conversion.
Create my voice →Your voice, ready for longer stories
Clone your voice. Let every document sound like you.
Record or upload a clean sample, create a reusable AI voice, and choose it when converting documents to audio.
- 10–30 sec recording
- Multilingual output
- You control deletion
What is AI voice cloning?
Voice cloning learns the recognizable qualities of a short recording—tone, rhythm, and character—then uses them to read new text. DocsToAudio connects that voice to the same document-to-audio workflow you already use.
02
From one sample to a reusable narrator
A guided four-step flow keeps recording, consent, preview, and use in one place.
Record or upload
Read the suggested passage for 10–30 seconds, or upload a clear MP3 under 10 MB.
Confirm permission
Confirm that the voice is yours, or that you have explicit permission to clone it.
Create and preview
Name and create your cloned voice, then generate a short preview to check the result.
Convert long documents
Choose the cloned voice in the voice picker and convert a long document, multiple chapters, or an entire book in one workflow.
03
One home for multiple cloning models
Once created, cloned voices appear in the document-to-audio voice picker. Switch between them freely and use your chosen voice for long documents, multiple chapters, or complete books.
MORE MODELS
More voice cloning models
More voice cloning options are planned, giving you additional choices based on voice quality and your production needs.
04
Who is voice cloning for?
For people who want a consistent, familiar narrator across the content they create, study, or publish.
Authors and creators
Turn drafts, essays, and member content into audio with a consistent narrator.
Learning and review
Listen to study notes and long documents in a voice that feels familiar.
Accessible publishing
Offer another way to experience written material without recording every chapter manually.
05
Why use DocsToAudio voice cloning?
Create the voice once, then use it naturally throughout the same document-to-audio workflow.
Use the voice you actually want
Create audio with your own voice or another specific voice you are explicitly authorized to use, instead of being limited to preset voices.
Convert long documents in one workflow
Convert long documents, multiple chapters, or an entire book without generating many short clips and stitching them together manually.
Switch voices inside the conversion flow
After creation, your cloned voice appears in the document-to-audio voice picker, where you can select and switch it just like any other available voice.
06
Consent and control are part of the flow
Voice data is sensitive. The current production flow sends the sample to ElevenLabs to create and host the cloned voice. DocsToAudio records your consent and gives you controls to preview and delete the voice.
- Only clone your own voice or one you are explicitly authorized to use.
- Deleting a voice is permanent and the original charge is not refunded.
- Do not use cloned voices for impersonation, deception, or unlawful content.
Voice cloning questions
How long should my recording be?
Use a clear recording between 10 and 30 seconds. Consistent speech without background noise matters more than filling the full time.
Which files can I upload?
The current upload flow accepts an MP3 file up to 10 MB. You can also record directly in a supported browser.
Can I clone another person’s voice?
Only when you have their explicit authorization. You must confirm permission before creation, and impersonation or deceptive use is prohibited.
How is voice cloning charged?
Each cloned-voice slot costs 5,000 credits, and each account can hold up to two active slots. The slot fee is charged once when the voice is created and is not refunded after deletion. Document conversion is charged separately at the same usage rate as standard ElevenLabs voices.
Can I use the voice in other languages?
You can use it with multilingual text, but similarity and pronunciation vary by model, sample language, and target language. Preview the result before a long conversion.
DOCS → VOICE