<ul><li>ElevenLabs has launched its new flagship text-to-speech model, Eleven v3, featuring inline audio controls, dialogue generation, and multi-speaker support for over 70 languages.</li><li>The model introduces audio tags for emotional control, supports dynamic speech turns with natural interruptions and shifts, offering significant advancements over previous versions.</li><li>The release emphasizes the improved audio quality and expressive nuances of v3, enhancing cadence, stress, and emotion across different languages.</li><li>While v3 is not recommended for real-time or conversational use due to latency issues, a promotional 80% discount on UI-based usage is offered until June, with public API access expected in the future.</li></ul>

ElevenLabs Unveils ‘v3’, Its Most Expressive Text-to-Speech Model Yet

Discover more