Get startedQuickstart

Voice Cloning API

Create a voice from a short sample, then synthesize with it

Voice cloning creates a synthetic version of a specific voice from a short sample, then synthesizes any text in that voice. It is part of the Build API: you create a voice, get an ID, and use it on the same speech endpoints as any catalog voice.

Creating a voice requires a consent record with the speaker’s full name and email. There is no create path that skips it, so a voice is only ever cloned with authorization.
1

Set your API key

$export SPEECHIFY_API_KEY="your-api-key-here"
2

Prepare a sample

Record 10-30 seconds of clean speech, under a minute and under 5MB, with no background noise. Sample quality is the biggest factor in the result.

3

Create the voice

Send the sample and a consent record to POST /v1/voices as multipart form data:

POST
/v1/voices
1curl -X POST https://api.speechify.ai/v1/voices \
2     -H "Authorization: Bearer <token>" \
3     -H "Content-Type: multipart/form-data" \
4     -F avatar=@<file1> \
5     -F consent="string" \
6     -F gender="male" \
7     -F name="string" \
8     -F sample=@string
Try it

The response returns a voice_id.

4

Synthesize with the cloned voice

Pass the voice_id to POST /v1/audio/speech exactly like a catalog voice:

POST
/v1/audio/speech
1curl -X POST https://api.speechify.ai/v1/audio/speech \
2     -H "Authorization: Bearer <token>" \
3     -H "Content-Type: application/json" \
4     -d '{
5  "input": "Hello! This is the Speechify text-to-speech API.",
6  "voice_id": "geffen_32",
7  "audio_format": "mp3",
8  "model": "simba-3.2"
9}'
Try it

Required fields

FieldRequiredNotes
nameYesA label for the voice.
genderYesmale, female, or not_specified.
sampleYesThe audio sample (binary).
consentYesJSON string containing fullName and email.
localeNoDefaults to en-US.

Model support

Cloned voices run self-serve on simba-3.0, simba-english and simba-multilingual. Use simba-multilingual to speak a cloned voice across 30+ languages from one voice ID.

simba-3.2 also serves cloned voices, currently as a limited release enabled per workspace — contact Speechify to have it enabled. Each voice’s models array in GET /v1/voices reflects what your workspace can actually synthesize, so branch on that rather than assuming. simba-3.2 is English only: a cloned voice with a non-English locale returns 400 there.

Manage voices

ActionEndpoint
Get a voiceGET /v1/voices/{voice_id}
Download the sample (WAV)GET /v1/voices/{voice_id}/sample
Delete a voiceDELETE /v1/voices/{voice_id}

Next steps

Voice Cloning Quickstart

The shortest path to your first cloned voice.

Text-to-Speech API

Synthesize with any voice, cloned or catalog.