Tiny Desk AI Video Generator
How it worksYour photos. Your friends. Your tiny concert. Pick your band, choose your look, and turn a favorite portrait into an 8 or 19-second performance worth sharing.
Who’s in your band?
Click a performer to replace them. Keep the rest of the band.
Replace 1–5 people. Clear, single-person photos · JPG / PNG / WebP · up to 10 MB each.
Every model receives the performance video and the five assigned portraits. Credits update with your selection.
Advanced settings
Uses the reference performance and matching template soundtrack.
190 credits · Replace at least one performer to start
Watch the reference performances
Reference examples from TinyDesk for the cast, shots and performance. These are not Pixwit-generated results.
A little desk. Your whole band.
Step into the band with Tiny Desk AI. Turn a portrait into a playful office-style concert, put a friend on lead vocals, or bring your whole group into the lineup. Choose each performer yourself, from the three backing singers to the lead and saxophonist, and let the performance reference guide the scene.
Make a quick 8-second highlight or choose the 19-second performance with a saxophone solo. Pick the AI model, picture settings and optional template soundtrack that suit your idea, see the credit quote before you generate, then preview, download and share the finished clip.
Three steps. One little concert.
01Cast the band
Choose a performer and upload a clear portrait. Remove it to restore that original band member.
02Make it a performance
Choose a model, length, picture settings and soundtrack option, then review the credit quote. Your portraits and the performance reference guide the generated shots; optional music is added afterward.
03Take it to the group chat
Preview your finished video, download it, and give the group chat something to talk about.
A few things to know.
Can I make a Tiny Desk AI video with one photo?
Yes. Upload a clear single-person photo for any backing singer, the lead singer or the saxophonist. Unselected roles use the original performer references. Add up to five photos to cast the whole band.
What is the difference between the 8 and 19-second versions?
The 8-second version follows the backing-singer shot and lead close-up. The 19-second version adds a saxophone solo and a return to the backing singers. Replacing the saxophonist requires 19 seconds. Each version offers compatible models and their supported settings.
How do I get the best performance from my photos?
Start with a clear, evenly lit single-person portrait for each role. Try the available models and resolutions to find the look you like. As with AI-generated video, character, motion and background details may vary.
Does this include the original Outstanding song?
The performance video guides the shots and motion. With Include template soundtrack enabled, the matching soundtrack is added after silent-video generation. Turn it off for a silent result. Adding the soundtrack does not establish precise lip sync.
How much does it cost? Is there a free preview?
The total credit quote below the button includes the selected model and reference-video workflow and soundtrack processing. Sign in with enough credits to generate. A separate free generation preview is not currently offered.
What video can I download?
Create an 8 or 19-second video with the selected model’s aspect ratio and resolution. Play and download the finished clip here, or find it in your generation history.
Can I come back while my video is generating?
Yes. Sign in and reopen this page to restore the latest task. Progress updates automatically. A retry button appears for failed steps that can be resumed.