Hedra takes a still image and a bit of audio (spoken words or song) and generates a really impressive lip sync video clip. Hedra doesn’t just add moving lips, it adds movements in the whole face, the eyes, the head, and the upper body. It’s really impressive. Perfect? No. But neither am I.
Using Hedra to create a lip sync is very easy and intuitive. No confusing settings or option-overload. Hedra uses a simple recipe for creating the lip sync video: Audio + Character = Video.
You can upload audio that you’ve recorded or generated in another platform, like ElevenLabs. You can also just type in the text you want, pick an AI voice, and generate the audio right within Hedra. If you want to make a character sing, import a recording of yourself singing or a song you generated in something like Suno.
After you’ve settled on the audio you want to you use, create a character. The character is an image of someone who will be doing the talking or singing. You can describe your character right within Hedra and it will generate an image for you. Or, you can import an image of yourself or something you generated on another platform, like Leonardo AI.
In about a minute (in my experience) your image will be speaking or singing the audio, complete with the face, head, and body movements that make sense and look realistic.