Make a photo talk or sing with your own audio
Talking & Singing Avatar turns one photo and one audio file into a video of that face speaking or singing. It is a photo to video AI avatar tool: you bring the picture and the sound, and the audio stays exactly as you uploaded it, so there is no dubbing or translation step. Use it when you have a picture rather than footage of someone speaking.
Examples
Talking avatar in English
AI-generated example
Talking avatar in Arabic
AI-generated example
Talking fox character in English
AI-generated example
Singing photo
AI-generated example
Singing fox character
AI-generated example
How to make a talking photo
- 1.
Upload a photo
Choose a JPG, PNG or WebP picture with a clear face, up to 20 MB.
- 2.
Add your audio
Upload speech, music or a song as an audio file, from 1 second to 3 minutes. The audio is used as it is and sets the length of the video.
- 3.
Pick the quality and start
Choose Standard 480p or High Quality 768p, see the credit cost, confirm you have the right to use the photo and the audio, and start. The finished video appears in My Projects.
What you can use it for
Talking photo
Give a portrait a voice: a spokesperson, a character or your own picture reading a script you recorded.
Singing photo
Make a photo sing by uploading a song or a vocal track. The audio can be music as well as speech.
Promos and social posts
Turn a spokesperson picture or a brand character into a short video for an announcement, an ad or a social post.
Lessons and training
Let a presenter photo narrate a lesson, a course intro or a training clip when filming is not practical.
Price
Talking & Singing Avatar uses your Voixie credits. Standard 480p costs 20 credits and High Quality 768p costs 40 credits for every started 15 seconds. One video can use up to 3 minutes of audio. The page shows the cost before anything is charged, and credits are refunded automatically if the video fails. Free accounts share the Lip Sync limit of 60 seconds, paid with their credits.
Talking & Singing Avatar or Lip Sync?
Lip Sync changes an existing video: it needs a person visibly speaking with their lips moving, their original voice and new audio, and it redraws the lips to match. Talking & Singing Avatar starts from a still picture, so nobody has to be filmed. If your video has a speaking face, use Lip Sync. If you have a photo, or a video that does not move, use Talking & Singing Avatar.
Questions about talking and singing photos
How do I make a photo talk?
Upload a photo with a clear face and an audio file of the speech, choose the quality and start. The video is made from those two files, with no filming.
Can I make a photo sing?
Yes. Upload a song or a vocal track instead of speech. There is no dubbing here, so the audio you upload is the audio you hear in the video.
Which languages does it work with?
You do not choose a language here. The video follows the sound of the audio you upload, and nothing is translated. To translate a video into another language, use the dubbing tool.
How long can the video be?
The audio can be from 1 second to 3 minutes, and the video runs as long as the audio. The photo can be up to 20 MB.
Whose photo can I use?
Only a photo and audio you have the right to use. Before a video starts you confirm this, and if the photo shows a child you confirm you are their parent or guardian or have their consent. Uploaded pictures go through our content policy check.