


Not a still or a waveform behind your MP3: a photo lip synced to your audio.
A voiceover, a podcast clip or a voice note from your phone. Common audio files such as MP3, WAV and M4A work.
Pick a voice and type the script, or clone your own from a short clean sample at 2 credits a minute of audio.
On the fast engine the median 17 second clip takes about 86 seconds to render.
episode-clip.mp3

One image and one audio file in, an MP4 of that face speaking out.

The face moves with every word of the audio. This clip is a real InfiniteTalk Fast output from one still image and one audio track.
0:00 to 0:1524 credits on the fast modelYou pay per second of video, so cut a long episode down to the part you want to post. Clean speech with one person talking at a time works best.
720pThe fast lip sync model renders 480p at 2 credits a second. The higher quality model renders 720p at 6 credits a second, or 480p at 3 as InfiniteTalk in the Playground (4 as High quality in the app).

If you only want a still image or a moving waveform behind an MP3, our audio to video converter tool does that in your browser with no sign up.
Open the converter

Cut the moment you want to share and give it a speaking face.

A recorded voiceover becomes a presenter speaking it.

Upload the recording of the lesson and one photo of the teacher.

A voice note from your phone, spoken by your own photo.

No camera, no editing timeline.
Upload a voiceover, a podcast clip or a voice note. No audio yet? Type a script and pick a voice instead.
portrait.jpgOne clear, front facing photo: you, a presenter you have permission to use, or a character you made.
The mouth, jaw and head move with the audio, and you download an MP4. A 10 second clip is 20 credits on the fast model.

Credits work out at about 2 cents each depending on the plan. Plans start at $12 a month; see the pricing page.
Upload your audio file and one photo of a face, and Percify lip syncs the face to the audio and returns an MP4 of that person speaking it. The playground quotes the exact cost before anything runs.
It is the AI kind: a face speaks your audio. If you only need a still image or a moving waveform behind your MP3, use our free audio to video converter instead. It runs in your browser, needs no sign up and adds no watermark.
Yes. Upload the MP3 with a photo and you get an MP4 of the photo speaking it. Other common audio files such as WAV and M4A work too.
The fast lip sync model costs 2 credits a second of audio, so a 15 second clip is 30 credits and a 60 second clip is 120. The higher quality model is 3 credits a second at 480p as InfiniteTalk in the Playground (4 as High quality in the app) and 6 at 720p on both. Runs that fail are refunded.
On the fast engine the median 17 second clip takes about 86 seconds to render. The higher quality engine takes longer.
Yes. Any speech works best when one person is talking at a time and the recording is clean. Cut a long episode into the clip you want to share, because you pay per second of video.
Yes. Pick a voice and type the script, or clone your own voice from a short clean sample at 2 credits a minute of audio, then make the photo speak it. Only clone a voice you have the rights to.
Your own, or someone who has agreed to it. Percify’s terms require that you hold the rights to what you upload. Do not upload a public figure or anyone who has not agreed.
No plan adds a watermark, and under the terms you own the content you create, so you can use it in ads, courses and client work. Output is not exclusive.
One photo and a script is all it takes.
Cancel anytime