Easily lip-sync AI Avatars to deliver your brand message

Lip-Sync is a generative AI process where the AI Avatar video is regenerated to speak the audio script you create. The AI Avatar will have natural, hyper-realistic facial expressions and lip movement as it speaks. To Lip-Sync an AI Avatar, you first place an AI Avatar on the video canvas scene. You then click on the Scene Voice button on the video canvas timeline. The Scene Voice Panel opens up and shows you a place to enter the script for each scene in your video. Enter in the script for the scene where you have placed the avatar, press the Select Voice button and pick a voice you want the avatar to have. Next, click the Generate Audio button. This will create the audio using your script. The audio is automatically added to the scene. Once you have created the audio you can return to the video canvas and press the Lip-Sync Actors button whenever you are ready to create the final video. The Lip-Sync process can take around 2-3 minutes to complete, and will return your avatar with the new voice and reading your script.

Yes, in fact it is very easy to create your own custom AI Avatar by uploading a photo of yourself looking at the camera. You can use the Create Greenscreen AI Avatar button and make your own avatar in around 2 minutes. Once you have your avatar created, you can give it your own voice using the Scene Voice Panel. Click on the Select Voice button, then under the Brand Voices tab, choose to Create Your First Voice. Record the script you are shown (30 seconds of reading) and this will make a digital voice actor that sounds like you. In the future when you are creating videos of your own AI Avatar, simply select your custom brand voice.

You can create lip-sync scripts that last minutes long. The longer the script, the more time it takes for the final lip-sync video to generate. The cost is determined by the length of the lip-sync you make. Remember, you will want to make sure the AI Avatar you choose is still in the scene and remains looking at the camera or in a stationary position if you are going to have longer lip-sync videos. When a lip-sync script lasts longer than the AI Avatar video, the video will loop back to the beginning as it generates. This can result in the video jumping from where it was with the avatar talking, back to the start frame of the video. In many cases this is fine, but depending on your AI Avatar, you may want to shorten the script to 10 seconds or less.

We offer 29 languages to choose from. It is super easy to create new scripts in any language as we have a built in translation feature that uses AI to convert your script from the original language into the target language you need. Click on the Scene Voice button on the timeline, and the Scene Voice Panel opens up. Enter in the scripts for each scene, then choose the language you want from the Translate All Scripts menu at the top of the panel. It takes, on average, around 15 seconds to translate a set of scripts. You will see the new script appear in the chosen language. Click the Generate Audio button for each script when you are ready. There is also a fully automated Translate Video Project option that will translate all text elements and use the current voice actor to regenerate the scripts automatically in the chosen language. You can find this option on the main video canvas timeline. Look for the Orange Globe button on the lower left corner of the timeline (Translate all text and audio). Press this button and a modal will show up giving you the options to automatically translate all text and audio on the project. All scripts will be translated and the new audio regenerated using the selected actor.

Yes, in fact this is the most common method for making training videos or video presentations using AI Avatars. Place the same AI Avatar or a variation of them (we offer Looks for select avatars which are different variations of the same actor such as standing or sitting) on each scene. Enter in a new Scene Voice script for each scene and the actors will read your lines in the selected voice. You should always choose the same actor voice to keep things consistent across your video scenes.

If you are using a Green Screen AI Avatar, you can place the avatar over elements on the scene such as another video, images, text or add shapes behind them. The Bake Elements option will render the AI Avatar, using the green screen, as a composite over the top of the elements. All elements on the scene become part of the composite video. We then remove all elements from the scene (as they are now part of the final video) and return the generated lip-syc as the background video on the artboard. This is a powerful feature for making actors that have design elements behind them. You can create a b-roll video, set it as the artboard background (right click on the b-roll video and choose Make Artboard Background) and then place a Green Screen AI Avatar over the b-roll footage. The new Lip-Sync generation will composite them together into a single video.

The Lip-Sync feature, while very powerful, does have one inherent limitation. If the script you enter in exceeds the length of the AI Avatar video, the lip-sync starts at the beginning of the video which in some cases, will make the video appear to jump as the actor speaks. For example, if you have an AI Avatar in a scene walking towards the camera, and create a very long lip-sync script, the actor will appear to jump back to the original position and begin walking forward again. This limitation can be worked around using one of three techniques. First, create shorter scripts to deliver the message and place the actor on different video scenes (duplicate the scene) to deliver each set of lines. The second technique is to use an AI Avatar that is stationary and has no other moving elements in the scene. The Green Screen AI Avatars are make to always face the camera and have limited hand gestures to minimize video jumps if a script is longer. The last technique is to use custom footage where you supply your own video of an actor talking for longer lengths of time.