ElevenLabs’ new v4 speech model supports more expression control and 90 languages

ElevenLabs has released its v4 speech model, which offers improved expression control and supports 90 languages. The model can clone voices with just a 10-second clip. The new model provides more nuanced and expressive speech synthesis, allowing for a wider range of emotions and intonations. This means developers and users can create more realistic and engaging audio content. ElevenLabs' v4 model is a significant upgrade to its previous version, which supported 60 languages and required a 30-second clip to clone a voice.