ElevenLabs’ new v4 speech model supports more expression control and 90 languages ElevenLabs v4 can clone voices with a 10 second clip By Ivan Mehta · TechCrunch ElevenLabs v4 can clone voices with a 10 second clip