Lip sync can be hard. We fixed that 💥 Introducing Speak on Freepik Upload a visual, add your own audio or a script, and get a lip-synced talking video in seconds → Custom voices in 30+ languages → Up to 5 min talking video → A lip sync tool,... made simple Available nowshow more

Go to Video > Tools > Speak Or click the link below 👇

Generate UGC clips that feel real, grab attention, and drive action Pick any image (illustrated, AI-generated, or real), write the words, and the tool handles the rest

A photo of a model turns into a localized talking ad in any language Add the script and no more filming video ads from scratch Scale your campaigns with Spaces and don't worry about budget

Your product photos can now present themselves Upload a shot, add your copy, and get a demo in any language your customers speak

All these are now ready to use in Spaces if you want to test out these workflows Speak will soon be in Spaces, but start working with audio nodes now

This looks amazing 😍 Proud to be partners in this and excited for what’s ahead🤝

That’s the energy we like to see!

Upload visual → add script → instant talking video. AI video creation keeps getting simpler.

@BeyondAI01 More time to let creativity flow 🔥

Great results!

super nice

Love hearing that, Joe 🙂

nice..

Glad you like it!

Has anyone tried it on music videos?

Awesome. Lip sync has always been the hardest part of AI

How about lip sync for a music video

At the moment the tool is focused on conversations

How does it hold up to singing? Is the camera controllable?

Its a super

Now go make your generations talk 🔥

But how to use it, this tool is a limited edition

Tools like this are changing how content is made. A single image + a script = a full talking video. We’re moving into a world where imagination matters more than equipment.

This is awesome

Let’s speak!! Amazing add Freepik crew

Time to give those visuals a voice

Awesome, what model is this using under the hood. Veed Fabric?

It uses Veed Fabric 1.0 and Sync React-1

Awesome! How does the template or agent know when to switch models?

Its speaking time, just try it and its awesome

Say it louder for the people in the back 🔥

Yay 🙌 finally we all can SPEAK now , tysm team for your great updates 🫶

"30+ languages and 5-minute videos? That's a massive leap for localized storytelling! Can't wait to see how creators leverage Speak to bridge the communication gap. Huge congrats on the launch! 🎉"

Tools that save time are truly great. 手間を減らすツールは本当に良いですね。

What model goes with speak? New LTX?

It’s built with Veed Fabric 1.0 and Sync React-1

This looks really handy. Lip syncing used to take forever to get right.

We felt the same pain!

Excellent work, it's great.

Go make some noise with it 🔥

Great tool!

We like your taste

😁

I hope we aren't being toyed with here. I know @Gossip_Goblin also is looking for good lip sync.

This looks great, gonna have to test it out but we still don't have the tools to both create structured lip syncing coupled with a camera move just yet. The first model to crack this will get alot of eyeballs. @Kling_ai 3.0 Omni comes the closest so far

5 mins is a massive unlock. most tools tap out at 30s before the sync breaks. partnership with veed makes this feel super native. clean.

Does it work for dialogues between two people?

Do you have an option of a Video as input for lip-syncing, not image?

I tested it and found the cost quite high, but the quality is still worse than InfiniteTalk for both images and videos. Lip-sync in images is fairly good, but unlike InfiniteTalk, it does not generate natural background motion. The video-to-video feature also has many errors.

epic!

If artificial intelligence is capable of simulating 'emotions' in design to the point that you cannot distinguish it from human work, does the 'human touch' still exist at all, or is it just an illusion we give ourselves?

What's the name of original software?

Hi @freepik !! I just spent 8.840 credits for this clip. Is this the expected result?