Voice Cloning

Use the MiniMax voice cloning capability to clone a voice and audition it with specified text. It stays consistent with MiniMax’s official /v1/voice_clone and is forwarded by GregAPI — no self-signing required.

export BASE_URL="https://api.gregapi.com"
export TOKEN="oh-xxxxxxxxxxxxxxxx"

Authorization: Bearer <TOKEN>
Content-Type: application/json

Endpoint

Operation	Method	Endpoint
Voice cloning	`POST`	`/minimaxi/v1/voice_clone`

Media type: application/json. Use cases: IP voice cloning, timbre cloning, and other scenarios that need quick voice replication.

Quickstart: clone a voice in three steps

1. Upload the audio to clone and get a `file_id`

curl -X POST "https://api.gregapi.com/minimaxi/v1/files/upload" \
  -H "Authorization: Bearer $TOKEN" \
  -F "purpose=voice_clone" \
  -F "file=@/path/to/audio.mp3"

Pass the returned file.file_id to POST /minimaxi/v1/voice_clone; file.filename is the original filename.

2. (Optional) Upload a sample audio to get the `prompt_audio` `file_id`

curl -X POST "https://api.gregapi.com/minimaxi/v1/files/upload" \
  -H "Authorization: Bearer $TOKEN" \
  -F "purpose=prompt_audio" \
  -F "file=@/path/to/short-sample.mp3"

3. Call the clone endpoint to finish cloning and get an audition audio

Request parameters (aligned with MiniMax official `/v1/voice_clone`)

Parameter	Type	Required	Description
`file_id`	string / integer	Yes	Audio `file_id` of the voice to clone, obtained from `/minimaxi/v1/files/upload`
`voice_id`	string	Yes	Target `voice_id` of the cloned voice; must meet the official naming rules
`clone_prompt.prompt_audio`	string	No	`file_id` of a sample audio (uploaded with `purpose=prompt_audio`)
`clone_prompt.prompt_text`	string	No	Text corresponding to the sample audio
`text`	string	No	Audition text (official limit ≤ 1000 characters)
`model`	string	Conditional	Required when `text` is provided; the TTS model for audition, e.g. `speech-2.8-hd`, `speech-2.8-turbo`
`language_boost`	string	No	Minor-language / dialect boost, e.g. `Chinese`, `English`, `auto`
`need_noise_reduction`	boolean	No	Enable noise reduction
`need_volume_normalization`	boolean	No	Enable volume normalization
`aigc_watermark`	boolean	No	Append an AIGC watermark rhythm at the end of the audition audio

Return fields

Field	Description
`input_sensitive`	Optional; returned when the input audio hits risk control, with `type` etc.
`demo_audio`	Returned when `text` and `model` are provided; the audition audio link
`base_resp.status_code`	Status code, 0 for success
`base_resp.status_msg`	Status description

Example: with audition audio

curl -X POST "https://api.gregapi.com/minimaxi/v1/voice_clone" \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "file_id": 1234567890,
    "voice_id": "MiniMax001",
    "text": "Artificial intelligence has been developing rapidly in recent years...",
    "model": "speech-2.8-hd",
    "need_noise_reduction": true,
    "need_volume_normalization": true
  }'

Billing

Billed by actual cloning and audition usage; see the MiniMax speech entries under “Model Pricing” in the console for exact prices.
Actual amounts follow the upstream consumption logs.

Large Language Models (LLM)

Image Models

Video Models

Speech Models

Endpoint

Quickstart: clone a voice in three steps

1. Upload the audio to clone and get a `file_id`

2. (Optional) Upload a sample audio to get the `prompt_audio` `file_id`

3. Call the clone endpoint to finish cloning and get an audition audio

Request parameters (aligned with MiniMax official `/v1/voice_clone`)

Return fields

Example: with audition audio

Billing

​Endpoint

​Quickstart: clone a voice in three steps

​1. Upload the audio to clone and get a file_id

​2. (Optional) Upload a sample audio to get the prompt_audio file_id

​3. Call the clone endpoint to finish cloning and get an audition audio

​Request parameters (aligned with MiniMax official /v1/voice_clone)

​Return fields

​Example: with audition audio

​Billing

Endpoint

Quickstart: clone a voice in three steps

1. Upload the audio to clone and get a `file_id`

2. (Optional) Upload a sample audio to get the `prompt_audio` `file_id`

3. Call the clone endpoint to finish cloning and get an audition audio

Request parameters (aligned with MiniMax official `/v1/voice_clone`)

Return fields

Example: with audition audio

Billing