Technology TechCrunch · 2026-09-28T14:00:00Z · 1 min read ElevenLabs’ new v4 speech model supports more expression control and 90 languages ElevenLabs v4 can clone voices with a 10 second clip Read original on TechCrunch