BLAST RADIUS

Frontier Line and access · releases

Gradium AI releases a low-latency multilingual text-to-speech model

Sep 1, 2026, first seen via MarkTechPost

Gradium AI released a default TTS model reporting an 81.0% human-rated pass rate on 500 hard sentences across five languages and 216 ms median time-to-first-audio.

Read the original at marktechpost.comOpens the publisher's site in a new tab

More in Frontier Line and access