MiniMax drops Speech-02 with 200k-character TTS
MiniMax just launched Speech-02, a new text-to-speech model capable of processing up to 200,000 characters in a single prompt from any file or URL. It supports over 30 languages, unlimited voice cloning, and sub-second streaming, aiming directly at long-form audiobook and podcast generation.
MiniMax just launched Speech-02, an upgraded audio model designed to handle massive text-to-speech workloads in real time.
Long-form audio in one go
The flagship feature is its massive 200k-character input limit. Users can turn an entire document or webpage URL directly into a full-length audiobook or podcast without manual chunking.
On top of the expanded context window, Speech-02 supports over 30 languages, sub-second streaming latency, and unlimited voice cloning capabilities.
Why it matters
Generating long-form audio has always been a painful, fragmented workflow requiring scripts to be spliced, generated, and stitched back together.
MiniMax's 200k-character ceiling makes single-prompt audiobook creation feasible right out of the box. We think the battleground in TTS is shifting from mere voice realism to sheer throughput and streaming speed, though unlimited cloning will undoubtedly keep safety and moderation teams on high alert.
Sources (1)
- x.com(opens in a new tab)Originalx.com