ElevenLabs launches new speech-to-speech models for real-time AI conversations
ElevenLabs has introduced two new 'Eleven Realtime' speech-to-speech models engineered for faster and more accurate conversational AI. The v1 Max variant achieves a notable 98.8 percent score on the Big Bench Audio benchmark with a quick 0.57-second response time, setting a new benchmark for speed and precision in voice technology applications.
Source: The Decoder