speech-2.6-hd
RecommendedRecommended
speech-2.6-hd
MiniMax voice cloning
Build a reusable voice from an MP3 sample, then generate speech for stories, videos, and other creative projects. Manage saved voices and review generated audio in the Start The Vision workspace.
About this workflow
Start with a clear MP3 voice sample between 30 and 180 seconds, then create and name a voice clone. Saved voices stay available in your workspace for later speech generation.
Choose an available MiniMax speech model, set the language tag and delivery controls, and generate audio from your text. Use only voice samples you have permission to use.
Built for practical production
Create a clone from an MP3 sample between 30 and 180 seconds.
Save and reuse multiple voices in your production workspace.
Adjust available speech settings such as speed, pitch, and volume.
Generate speech in the languages currently supported by the studio.
Available models
Recommended
speech-2.6-hd
speech-2.6-turbo
speech-2.5-hd
speech-2.5-turbo
speed-02-hd
speed-02-turbo
speed-01-hd
speed-01-turbo
Model availability can change as the connected voice service updates its catalog.
How it works
Choose a clear MP3 between 30 and 180 seconds and confirm you have permission to use the voice.
Give the voice a recognizable name, save it, and choose it for your next generation.
Enter your text, choose an available model and language tag, adjust delivery, and listen to the result.
Frequently asked questions
The current studio accepts MP3 files up to 5 MB with a duration between 30 and 180 seconds.
Yes. Saved voices are available in your workspace so you can select them for later speech generations.
The model catalog currently lists speech-2.6-hd, speech-2.6-turbo, speech-2.5-hd, speech-2.5-turbo, speed-02-hd, speed-02-turbo, speed-01-hd, and speed-01-turbo. Availability can change.
The studio offers multiple language tags, including English, Chinese, Spanish, French, German, Hindi, Urdu, Japanese, and Korean. Check the language selector for the current full list.
Only upload and use voice samples that you own or have explicit permission to use, and follow applicable laws and service terms.