ElevenLabs Voice Design 2.0: Create Any AI Voice From a Text Description
ElevenLabs has released a major update to its Voice Design feature, allowing users to generate entirely unique AI voices from a plain-text description — no audio sample required.
What Changed in Voice Design 2.0
ElevenLabs has released a significant update to its Voice Design feature, making it substantially more powerful and predictable. The original Voice Design allowed users to describe a voice in text and receive a generated AI voice matching that description. Version 2.0 improves the fidelity of that generation — the gap between what you describe and what you receive has narrowed considerably.
Users can now specify age, gender, accent, speaking pace, emotional baseline, and tonal qualities in natural language. A description like "a calm, authoritative 45-year-old British man with a measured speaking pace and a slight warmth in his delivery" now produces a voice that demonstrably matches each of those parameters.
Why This Matters for Content Creators
For content creators building a consistent audio brand, Voice Design 2.0 solves a real problem. Voice cloning requires a sample of a specific person's voice. Voice Design generates a voice from scratch — meaning you can create a unique, brand-specific AI narrator that no one else is using, without needing a human voice actor to record a sample.
Podcast producers, YouTube channel operators, and e-learning course creators can now design a brand voice on paper and generate it in minutes. The voice is consistent across any amount of text, does not require royalties, and can be updated or refined at any time.
Availability and Pricing
Voice Design 2.0 is available on all paid ElevenLabs plans. Creator plan users ($22/month) receive full access to all Voice Design features including the updated generation engine. Starter plan users ($5/month) can generate and use Voice Design voices with their monthly character allowance.
ElevenLabs in 2026
This release continues ElevenLabs' pattern of shipping meaningful model improvements on a frequent cadence. The company has established a clear lead in voice quality over competitors, and updates like Voice Design 2.0 extend that lead into features that did not exist in other platforms at comparable quality levels. For content creators evaluating voice tools in 2026, ElevenLabs remains the obvious starting point.