AIMarkTechPostabout 6 hours ago
Gradium Launches Real-Time Speech Translation Models
11 min read
Gradium released stt-translate and s2s-translate, real-time speech translation models covering five languages and 20 pairs. The models stream results live in the browser and claim better accuracy and latency than gpt-realtime-translate and gemini-3.5-live-translate. Output voice control including cloning is also included.
Level
Hype check
Tap to vote and see what everyone thinks.
