MiniMax Speech
Speech-02 TTS with 10-second voice cloning and 30+ languages.
About
MiniMax Speech is the audio platform of MiniMax, the Chinese AI company founded by former SenseTime veterans. Its flagship Speech-02 text-to-speech model supports 30-plus languages, processes up to 200,000 characters, and offers real-time synthesis with emotion control, while voice cloning requires only about 10 seconds of reference audio.
The Speech-02 series includes Speech-02-HD for high-fidelity voiceovers and audiobooks, Speech-02-Turbo for low-latency real-time applications, and dedicated voice-cloning models, with 300-plus prebuilt voices, controls for pitch, speed, and volume, and cross-lingual synthesis. The model is ranked number one on the Artificial Analysis text-to-speech leaderboard according to its AWS Marketplace listing.
MiniMax Audio is accessible through the MiniMax API and platform, with a free tier for basic text-to-speech and credit-based paid plans. MiniMax raised a 600 million dollar Series B led by Alibaba in 2024 at a reported 2.5 billion dollar valuation.
What it can do
- Speech-02 HD and Turbo TTS models
- 10-second voice cloning
- 30+ languages with emotion control
- 300+ prebuilt voices
- API, SDK, and web platform access
Setup
Sign up at minimax.io/audio or use the MiniMax API with a Group ID and API key.
Prerequisites
A MiniMax account; API key for programmatic access.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 133; Trending score 82/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.