Tiny Desk AI FAQ

The Outstanding Tiny Desk AI trend is an AI-made version of the Tiny Desk meme, inspired by the Outstanding desk-concert performance. Our independent tool lets you choose portraits for 1–5 performers in a 15-second video with generated sound. Watch the published example; ordering is not open yet.

No. This is an independent AI video service, not NPR, NPR Music, Tiny Desk Concerts or an artist’s official website. The original concert link is provided for context, not as our generated example.

Choose any 1–5 of the five performers: front backing singer, left backing singer, right backing singer, lead singer and saxophone player. The white-capped performer at the front sings backing vocals and plays a tambourine; the red-capped lead singer appears in the middle close-up. Add one photo to each role you want to replace. Empty positions keep the original performers; the lead singer is optional. Photo selection order does not change the roles.

Choose a clear individual portrait with the face and distinctive features visible. Use JPG, PNG or WebP up to 10 MB. Avoid crowded images, heavy filters, sunglasses and motion blur. Real people must be consenting adults. Do not submit minors, celebrities, public figures or people who have not agreed to appear.

The 15-second output includes generated music, singing and saxophone performance guided by the reference. Listen to the published sample. It is not a promise of an exact original recording or voice cloning; sound and lip sync can vary.

There is no free personalized generation, subscription or credit pack. The planned price is US$4.99 per video. Orders are currently closed and no payments are being taken.

The video is 15 seconds, vertical 720 × 1280, with sound. Generation time is separate from video length. Measured delivery estimates and a latest delivery limit will appear before checkout opens. You can then close the page and find the result in My videos.

AI likeness, proportions, movement and timing can vary. The published example shows all five roles replaced; results for other combinations may differ. Compare the complete example before buying. Severe distortions or an unusable result are covered by our remedy rules.

Choose one automatic redo or a full refund. If the redo fails, you receive a full refund. Rejected photos can be replaced once for free or refunded.

It stays on this device until payment succeeds. Our stored photo copies are removed within 7 days of delivery, cancellation or refund; videos are available for 30 days. Processing providers have their own rules. See Privacy for details.

Sign in to the same account and open My videos. The delivery email links to your account. Keep your own downloaded copy before the 30-day download period ends.

No. The intended flow runs in your browser: choose a portrait for each selected role, then download the finished MP4 with generated sound. No custom prompt, timeline editor or separate editing app is required. Custom lyrics, voice uploads and voice cloning are not offered.

Yes. Add a portrait only to the role you want to replace and leave the other slots empty. Empty slots keep the reference performers. You can choose any 1–5 roles; the lead singer is optional.

No. Use a separate, clear individual portrait for each role you want to replace. Friends do not need to pose together. The tool does not automatically split a group photo or assign its faces to roles; choose each role and add its portrait yourself.

No. Each role appears during its part of the fixed performance. Watch the full example to compare the lead singer, three backing singers and saxophone player. Framing and face size differ between roles, and AI timing can vary; we do not promise equal screen time or a close-up for every person.

The planned download is a 15-second vertical MP4 at 720 × 1280 with generated sound and no watermark. Subtitles and custom captions are not offered. Orders remain closed; final purchase details will be shown before checkout opens.