minimax/voice-clone
Voice Clone is MiniMax's text-to-speech AI model. Generate human-like voiceovers with expressive intonation, multilingual support, and customizable voice characteristics.
Example output — click Run to generate your own
Send your first request
OpenAI-compatible endpoint with unified authentication and usage tracking.
https://api.sandbase.ai/v1/runminimax/voice-cloneInput Schema
5 parameters · 0 required · 5 optional
| Parameter | Type | Required | Description |
|---|---|---|---|
text | string | Optional | Text to generate a TTS preview with the cloned voice (optional) · Max length: 1000 · Default: "Hello, this is a preview of your cloned voice! I hope you like it!" |
audio | string | Optional | URL of the input audio file for voice cloning. Should be at least 10 seconds long. To retain the voice permanently, use it with a TTS (text-to-speech) endpoint at least once within 7 days. Otherwise, it will be automatically deleted. |
accuracy | number | Optional | Text validation accuracy threshold (0-1) · Min: 0 · Max: 1 |
noise_reduction | boolean | Optional | Enable noise reduction for the cloned voice · Default: false |
need_volume_normalization | boolean | Optional | Enable volume normalization for the cloned voice · Default: false |
Output Schema
| Field | Type | Description |
|---|---|---|
id | string | Unique identifier for the generation task |
status | string | Task status: pending, running, completed, failed, timeout |
model | string | Model used for the generation |
outputs | array | Array of output items |
outputs[].url | string | URL of the generated artifact |
outputs[].content_type | string | MIME type (e.g. image/png, video/mp4) |
error | object | null | Error details if failed, null on success |
error.type | string | Machine-readable error type code |
error.message | string | Human-readable error description |
Async Workflow
This model uses asynchronous execution. Submit a request and poll for the result.
- Submit — POST to /v1/run, receive an
id - Poll — GET /v1/run/{id} until status is
completed,failed, ortimeout - Retrieve — Read
outputsfrom the completed response
Code Examples
Ready-to-run snippets
# Step 1: Submit
curl -X POST https://api.sandbase.ai/v1/run \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "minimax/voice-clone",
"text": "Hello, this is a preview of your cloned voice! I hope you like it!",
"audio": "https://static.sandbase.ai/examples/minimax/voice-clone/input_audio_0.wav",
"noise_reduction": false,
"need_volume_normalization": false,
"prompt": "a beautiful sunset over mountains"
}'
# Step 2: Poll result (replace <id>)
curl https://api.sandbase.ai/v1/run/<id> \
-H "Authorization: Bearer YOUR_API_KEY"API README
MiniMax Voice Cloning
MiniMax Voice Cloning creates a reusable synthetic voice identity from a short reference recording. It extracts the speaker’s timbre and speaking character so later text-to-speech requests can reproduce a voice that remains recognizably connected to the source sample.
Clean source audio produces the most dependable result, but the route also provides noise-reduction and volume-normalization options when recordings need preparation. Optional transcript validation and preview text help assess similarity before the cloned voice is used in a larger narration workflow.
Highlights
- Rapid timbre replication. Builds a custom voice identity from a compact reference recording.
- Reference cleanup. Can reduce background noise and normalize level before extracting vocal characteristics.
- Transcript-assisted stability. Uses matching text when supplied to strengthen similarity and cloning consistency.
- Previewable result. Can return a sample reading so teams can evaluate the voice before broader synthesis.
Pricing
| Configuration | Billing unit | Price |
|---|---|---|
| Initial clone operation | Base charge | $1.50 |
| Preview text | Each started 1,000 characters | $0.30 |
When to Use
| ✅ Good fit | ❌ Consider alternatives |
|---|---|
| The project needs building a reusable voice from consented reference audio | The goal is a different media task or endpoint |
| The available inputs match the required local schema | Required source media or permissions are unavailable |
| The brief can specify subject, composition, style, and delivery | The result must be deterministic at pixel or sample level |
| The supported formats and controls match final placement | Delivery requires unsupported dimensions, codecs, or duration |
| An asynchronous generated result fits the workflow | A live, frame-synchronous, or real-time response is mandatory |
Prompt Guide
Write the request as a production brief: identify the main subject or source material, state the intended transformation, describe composition or timing, and finish with style, atmosphere, and delivery constraints. Keep preservation requirements separate from requested changes, and use only fields exposed by this route.
{
"text": "Hello, this is a preview of your cloned voice! I hope you like it!",
"audio": "https://example.com/reference.mp3",
"accuracy": 0
}
Technical Specs
| Spec | Value |
|---|---|
| Model ID | minimax/voice-clone |
| Input fields | text (string)<br>audio (string)<br>accuracy (number; 0–1)<br>noise_reduction (boolean)<br>need_volume_normalization (boolean) |
| Required input | None marked required |
| Output fields | url, content_type, custom_voice_id |
| Execution | Asynchronous job |
Related Models
minimax/speech/2.8/hdminimax/speech/2.8/turbominimax/music/v2