How to clone a voice
Clone a voice in four steps: sample, consent, create, synthesize.
- Record a 10-30 second clean sample
- Capture consent (full name and email)
- POST /v1/voices with multipart/form-data
- Synthesize with the returned voice ID
Definition
Cloning a voice is a four-step flow on the Speechify Build API: record a sample, capture consent, create the voice, and synthesize with it. Zero-shot cloning makes this self-serve, so the whole path fits into an app or a script.
The steps
First, record a clean sample of 10 to 30 seconds, under a minute and under 5MB, with no background noise. Second, capture the speaker’s consent: their full name and email. Third, POST to /v1/voices as multipart/form-data with the name, gender (male, female, or not_specified), the sample as binary, and the consent record as a JSON string. You receive a voice_id.
Synthesizing
Pass the voice_id to POST /v1/audio/speech, or the streaming endpoint, with your text. The cloned voice speaks it. Use simba-multilingual to reach 30+ languages; the curated simba-3.2 set plays cloned voices too, but each one currently requires manual Speechify approval.
Consent is required
The consent field is mandatory and must contain the speaker’s full name and email. There is no create path without it, so you cannot clone a voice you have not been authorized to clone. See the ethics page and the voice cloning API.
Frequently asked questions
How do I clone a voice?
What sample quality do I need?
Which models can play my cloned voice?
Start building
Clone a voice in four steps: sample, consent, create, synthesize.