AI Talking Photo
AI Talking Photo animates a still portrait with new speech. Turn a headshot, character, or illustrated face into a short speaking video for messages, explainers, and social content.
Upload a prepared audio track, or type a script and select an available language and voice. Preview generated speech before submitting, then use positive and negative prompts to guide the visual result.
Choose a clear face in a JPG, JPEG, PNG, or WebP image up to 20 MB. The image must be 300 to 5,000 pixels on each side with an aspect ratio from 0.4 to 2.5. Speech can run from 3 to 120 seconds.
It turns a still portrait into a video whose visible face speaks in sync with uploaded audio or generated text-to-speech.
Yes. Enter a script, choose an available language and voice, and preview the speech before generating the video.
Use a JPG, JPEG, PNG, or WebP image up to 20 MB. Images must be between 300 and 5,000 pixels on each side and use an aspect ratio from 0.4 to 2.5.
The current workflow accepts speech from 3 to 120 seconds and asks before trimming audio that exceeds the limit.