如果想在 V2EX 获得更好的推广效果,欢迎了解 PRO 会员机制:
https://www.v2ex.com/pro/about

如果你经常使用铜币置顶主题,持有 V2EX Solana Token 会在每日签到时获得额外铜币:
https://www.v2ex.com/solana
mesbbs
V2EX  ›  推广

Dime Dime Baby AI Video Generator: turn one photo into the viral Bby Wow dance (720P, original audio, no prompt)

  •  
  •   mesbbs · 4h 19m ago · 109 views

    Hi V2EX 👋

    I made Dime Dime Baby AI ( https://dimedimebabyai.site). You upload one photo and get the viral Dime Dime Baby / Bby Wow dance video back, with the person in your photo as the dancer. The output is 10 or 15 seconds, 720P, vertical 9:16, with the original song. No filming, no CapCut, no prompt writing.

    What is Dime Dime Baby AI?

    Dime Dime Baby AI is the AI version of a TikTok meme from October 2026. In the original clip, a man in a black cowboy hat walks out of the crowd at a quinceañera party, bites his finger, brushes past the camera, walks away and comes back to point at the viewer. Spanish-speaking TikTok calls it "Dime Dime Baby". English-speaking TikTok calls it "Bby Wow", after the Karol G, Judeline and Rusowsky song most versions use.

    Then people started replacing the dancer with AI: their friends, their dogs, cartoon characters. Searches for "dime dime baby ai" show up right after. There is no TikTok filter for this. Every AI version you see was made with an AI video generator or a CapCut template.

    How Dime Dime Baby AI works (3 steps)

    1. Upload one photo. One person, facing the camera, full body or waist-up works best.
    2. Pick a length. Choose 10 seconds (everything up to the final point) or 15 seconds (the full dance).
    3. Wait 5–10 minutes and download. The job runs on the server, so you can close the tab. The MP4 waits in "My videos".

    Why the result looks like the real meme

    I tested the output against the original one second at a time. Four things decide whether a Dime Dime Baby AI video works, and this is what I found for each.

    1. Same moves, same timing. The model doesn't invent a dance. It gets the original party clip as a reference video and only replaces the dancer. In my 10-second test, the walk-in, finger bite, close pass and turn all landed on the same timestamps as the original.

    2. Same person in every frame. My test photo was a guy in a leather jacket, white tee, silver chain and white sneakers. All of those stayed on him for the whole clip. The output never switched back to the original cowboy outfit, and the face never drifted into someone else's.

    3. Same party. The ballroom, the coloured lights and the guests filming on their phones are untouched. Only the dancer changes.

    4. The real song, in sync. The model renders a silent video, and the server then muxes the original soundtrack onto it without re-encoding the video. I decoded the audio from the final MP4 and compared it with the source track, and it was byte-for-byte identical. So you get no AI-generated music and no audio drift.

    The test render took about 8.5 minutes.

    Pricing

    Dime Dime Baby AI is paid per video, with no subscription:

    Length 1 video 3 videos 10 videos
    10 seconds $5.99 | $15.99 $49.99
    15 seconds $7.99 | $22.99 $74.99

    Every tier is 720P HD, vertical 9:16, with sound and no watermark. 10-second and 15-second videos are separate balances, so you never end up with a leftover half credit. Unused videos never expire. If a render fails or the safety filter blocks it, the video goes back to your balance automatically.

    Tech notes for V2EX

    • TanStack Start + React 19, deployed on Cloudflare Workers
    • Cloudflare D1 for data, R2 for photos, outputs and reference clips
    • A reference-video + image editing model, with the prompt and reference clips tuned for this one scene. The 10 s and 15 s versions each have their own reference clip cut to exactly that length.
    • Silent render, then the soundtrack is remuxed in the Worker with mediabunny
    • Stripe Checkout with an idempotent credit ledger, so webhook retries can't double-grant credits and refunds take credits back
    • A cron trigger every 2 minutes that finishes or refunds jobs after the user closes the tab

    Ground rules

    • Dime Dime Baby AI is an independent fan-made tool. It is not affiliated with Karol G, Judeline, Rusowsky, their labels, TikTok or CapCut.
    • Only upload yourself, or an adult who said yes. No minors and no celebrities.
    • Please mark your post as AI-generated where the platform asks.

    Quick facts

    • Name: Dime Dime Baby AI (also searched as "Bby Wow AI")
    • URL: https://dimedimebabyai.site, Spanish version at https://dimedimebabyai.site/es/
    • What it does: turns one photo into an AI video of the viral Dime Dime Baby / Bby Wow dance with that person as the dancer
    • Output: MP4, 720P HD, 9:16 vertical, 10 s or 15 s, original audio, no watermark
    • Time: about 5–10 minutes per video
    • Price: $5.99 (10 s) or $7.99 (15 s), from $5.00 per video in packs, one-time payment
    • Refunds: automatic for failed renders
    • Sign-in: Google
    • Languages: English, Spanish

    FAQ

    Is there a Dime Dime Baby AI filter on TikTok? No. TikTok and Instagram have no built-in Dime Dime Baby or Bby Wow filter. AI versions are made with AI video generators like Dime Dime Baby AI or with CapCut templates.

    Is Dime Dime Baby AI free? No. Every video runs on a paid AI video model. A 10-second video costs $5.99 and a 15-second video costs $7.99.

    Dime Dime Baby AI vs a CapCut template? Most CapCut templates need you to film yourself doing the moves. Dime Dime Baby AI needs one photo and puts you inside the original party clip.

    What does "dime dime baby" mean? It's Spanish for "tell me, tell me, baby", the line people hear in the song under the clip.

    Guides on the site:

    I'd love feedback, especially on likeness. If a result looks off, reply with what kind of photo you used and I'll look into it. Thanks!

    No Comments Yet
    About   ·   Help   ·   Advertise   ·   Blog   ·   API   ·   FAQ   ·   Privacy   ·   Solana   ·   3284 Online   Highest 6679   ·     Select Language
    创意工作者们的社区
    World is powered by solitude
    VERSION: 3.9.8.5 · 26ms · UTC 11:27 · PVG 19:27 · LAX 04:27 · JFK 07:27
    ♥ Do have faith in what you're doing.