Back to blog

What Is Talk to Me Doja AI? The Urara Shiraishi Cheek-Sway Trend Explained (2026)

Nov 17, 2026Talk to Me Doja Editorial TeamTalk to Me Doja Editorial Team

What is Talk to Me Doja AI? See how a 2015 Urara Shiraishi clip and Doja Cat audio sparked the Nov 2026 sway trend, and make yours from $4.99.

If your For You page has been filled with creators acting unbothered for two seconds before cupping both cheeks and swaying side to side, you are probably asking: what is Talk to Me Doja AI, and how are people turning still portraits into that exact dance?

The phrase refers to two closely connected things in November 2026:

  1. The viral TikTok meme and dance trend (often captioned "Speak to me Doja" or "Talk to me Doja"), which pairs an Episode 1 scene from the 2015 anime Yamada-kun and the Seven Witches with a sped-up edit of Doja Cat's "Like That".
  2. The single-photo AI motion generator at the Talk to Me Doja AI Video Generator, which transfers that five-beat cheek-cupping sway onto any clear front-facing photo to produce a 12-second, watermark-free 9:16 vertical MP4.

Below, we trace how an eleven-year-old anime body-swap gag turned into a multi-million-view TikTok format in four days, and explain how single-image motion transfer recreates the gesture without requiring you to film yourself.

The Four Stages of the Trend: From 2015 Anime to 2026 AI Remixes

To understand why the gesture looks so distinct, it helps to separate the four chapters of the trend's history:

StageDateCreator / SourceReachWhat Happened
1. Anime SceneApr 12, 2015

Yamada-kun and the Seven Witches Ep. 1

TV / Streaming

Ryuu Yamada (in Urara Shiraishi's body) cups her cheeks and sways when approached by Toranosuke Miyamura.

2. Meme SparkNov 7, 2026@lovingbalanci229,000 views (5 days)

Added caption "Speak to me Doja" over the anime loop with sped-up Doja Cat audio.

3. Live-Action BreakoutNov 9, 2026

@the_real.ryan / @bunnicutieluv

3.1M & 744K views (3 days)

Side-by-side human recreations of the cheek-cupping sway blew up on TikTok.

4. AI Video WaveNov 10, 2026

@twin.cities.mattress & AI tools

Global TikTok & Reels trend

Creators began animating single still portraits into the 12-second sway choreography.

The Joke Behind the Gesture

In Episode 1 of Yamada-kun and the Seven Witches (adapted from Miki Yoshikawa's manga by Liden Films), high school troublemaker Ryuu Yamada swaps bodies with quiet honors student Urara Shiraishi. Because Yamada has no idea how a polite girl reacts when Toranosuke Miyamura invites her to a club, he exaggerates a shy pose—cupping both hands under his chin, tilting his head, and wiggling his shoulders.

When @lovingbalanci added the caption "Speak to me Doja" on November 7, 2026 (Know Your Meme), the joke reframed that body-swap awkwardness: no matter how tough you act in everyday life, if Doja Cat spoke a single word to you, you would immediately fold into Urara Shiraishi's bashful cheek-cupping sway. We cover the full cultural history on our Talk to Me Doja Meme page.

How Single-Photo AI Motion Transfer Works

Before dedicated motion templates existed, joining a dance trend meant setting up a phone tripod, rehearsing the timing, and recording until you nailed the transition. General-purpose image-to-video tools were not much easier: typing "person cups cheeks and sways shyly" into a raw text-to-video prompt usually produced four-second clips with warped fingers or stiff shoulders.

Prompt-free motion transfer in the Talk to Me Doja AI Studio solves this by combining your source portrait with a pre-structured 12-second motion sequence powered by Kie's Seedance 2.0 video models (bytedance/seedance-2-mini and bytedance/seedance-2-fast):

  1. Reference portrait: The model uses one front-facing JPG, PNG, or WebP photo (up to 20 MB, at least 300 × 300 px) as a reference for the person and clothing. Details can vary in the output.
  2. Five-Beat Temporal Mapping: Instead of guessing what movement to invent, the engine steps through five fixed storyboard windows:
    • 0–2s (Cool Pose): Calm, unbothered expression facing the camera.
    • 2–5s (Cheek Cup): Both hands rise cleanly under the chin and cheeks.
    • 5–8s (Bashful Sway): Gentle head tilt and left-to-right sway with a shy pout.
    • 8–10s (Shy Glance): Quick look away, then a flustered glance back into the lens.
    • 10–12s (Camera Lean): Forward lean with a warm smile on the final beat.
  3. Optional AI music: Include an AI-generated backing track or export a silent MP4. For Doja Cat’s “Like That,” check your platform’s music library and follow the song guide.

Honest Product Boundaries: Pricing, Safety, and Privacy

We built talktomedoja.net around transparent rules so you know what to expect before buying credits:

  • One-time credit packs: Starter costs $4.99 for 15 credits, Creator costs $12.99 for 45, and Studio costs $29.99 for 120. Compare packs on the pricing page.
  • Four Render Options: Choose Fast 480p (15 credits), Fast 720p (22 credits), Premium 480p (30 credits), or Premium 720p (45 credits).
  • Credit returns: Generating a new clip uses credits. If the task is marked failed or rejected after a deduction, the credits used for that task return to your balance.
  • Media links: Use a photo of one adult who agreed to appear. Anyone with a working media link can open the file. See the Privacy Policy for details.

Want to see how to pick the best portrait and render your own video? Read our complete Talk to Me Doja AI Guide, check our step-by-step tutorial, or go straight to the Talk to Me Doja AI Video Generator to start creating.