- VEED's AI voice cloning now runs on Eleven v4, ElevenLabs' new speech model.
- v4 powers your cloned voice, not a library of stock voices. The voice in your videos is yours.
- Instant Voice Clones need around 10 seconds of audio. Professional Voice Clones are there for the highest fidelity.
- Your cloned voice remains recognizably yours across regenerations, long-form scripts, and 90+ languages.
- v4 is a new architecture that interprets tone, pacing, emotion, character, and context, so your script is performed rather than read.
- This feature is now available to any VEED user with voice cloning access.
VEED's AI voice cloning is now powered by Eleven v4, the new speech model from ElevenLabs. You record around 10 seconds of audio, VEED builds your voice clone, and every script you write after that is narrated in your own voice with real emotional range in more than 90 languages.
Here's the part worth saying out loud. Cloned voices have mostly failed on emotion, not on accent. They get the timbre right and the delivery flat, so the clone sounds like you reading a terms-and-conditions page. That's the problem v4 was built to solve.
What changed?
VEED's voice cloning now runs on Eleven v4. Eleven v4 is an entirely new model and architecture, not an increment on v3. It interprets tone, pacing, emotion, character, and context, so speech is performed rather than read.
In practice, that means a line you wrote as an aside sounds like an aside. Your product name lands with the emphasis you meant. The pause before the punchline is actually there.
Your voice, in 90+ languages
Speaker identity is consistent across regenerations, across languages, and across long-form work. That's the part that usually breaks, and it's why most people give up on cloning after the third take.
v4 covers 90+ languages with native cross-language delivery. ElevenLabs reports the biggest gains in Japanese, Brazilian Portuguese, Mandarin, and Cantonese. Narrate once, speak to every market you sell in, and still sound like yourself in all of them.
How to clone your voice in VEED
- Record around 10 seconds of audio for an Instant Voice Clone, or supply a longer sample for a Professional Voice Clone
- Write your script, or generate one with the script generator
- Generate the voiceover in your own cloned voice
- Make it yours: add subtitles, apply your brand kit, and publish
Who it's for
- Marketing teams of one: weekly social video narrated in the founder's voice, without booking the founder
- Founders: tell the product story yourself, in every market, without learning four new accents
- Business owners: update a script and regenerate, instead of re-recording the whole thing to fix one line
About VEED
VEED is the AI video creation platform that grows your brand on social. Imagine it, generate it, add your branding, and get post-ready video in minutes. Millions of marketers and founders use VEED, including 76% of the Fortune 500. VEED builds its own models, including VEED Fabric, and its APIs run AI video inside leading creative products.
Clone your voice once, and narrate everything after that. Start free, no card required.
