0:00 / 0:00

Emociones con IA 🥲 Hoy quise poner a prueba los mejores generadores de vídeo con IA para ver si de verdad son capaces de transmitir diferentes emociones 👀 Usé la misma imagen y el mismo prompt para generarlas, y aun así cada uno me da un resultado distinto… Os dejo los testeos que hice para que podáis juzgar vosotros mismos qué generador lo hace mejor 😋 Y, por cierto, mañana Kling lanza su nueva versión: Kling 3.0. Pronto tendréis nuevos vídeos poniéndolo a prueba Y como siempre, si comentas “ARIA”, te paso todos los prompts de las imágenes y de las emociones que usé 💌

Aria Cruz | Influencer AI

@soy_aria_cruz · Digital creator

INSTAGRAM · 2026-02-03Source

930likes

709comments

Remix This

Prompt

GLOBAL LOCK: A young Caucasian woman in her mid-20s with dark brown hair tied back in a sleek ponytail. She has fair skin with warm undertones, wearing large, round, thin-framed gold glasses. She is adorned with multiple layers of gold and pearl necklaces of varying lengths and thicknesses. She wears a gold sequined camisole. Large, highly detailed white feathered angel wings are attached to her back. A glowing, vibrant orange neon halo floats steadily above her head. The environment is a dark, moody indoor setting with soft, colorful bokeh in the background. The lighting is cinematic, with a strong warm rim light from the halo and soft fill light on her face. The color grade is warm and saturated with deep blacks. The camera is a high-end cinema lens (85mm feel) with shallow depth of field.

[00:00–00:03]
The woman is in a medium close-up, facing the camera directly. She has a frustrated and angry expression, furrowing her brows deeply and gritting her teeth. Her breathing is heavy and visible in her chest and shoulders. The camera is static. The glow from the halo casts a steady orange light on her hair and the tops of her shoulders.

[00:03–00:08]
The woman suddenly opens her mouth extremely wide and lets out a powerful, visceral scream. Her facial muscles tense significantly, her eyes squint shut with the effort, and her neck tendons become visible. The camera begins a subtle, high-frequency handheld shake to emphasize the intensity of the scream. The lighting from the halo flickers slightly as if reacting to the energy. The feathers on her wings vibrate slightly from the simulated sound. The transition into the scream is sudden and explosive.

NEGATIVE PROMPT: face melting, losing glasses, necklaces merging into skin, wings disappearing, halo flickering off, blurry textures, cartoonish look, low resolution, distorted eyes, extra limbs, static hair during scream, robotic movement, mouth not opening wide enough, flickering background.

SPEECH PACK:
Transcript: [00:03-00:08] (Visceral, high-pitched, guttural scream) "AAAAAAHHHHHHHHHHH!"
TAKE_A: High-pitched, glass-shattering scream with a sharp onset.
TAKE_B: Guttural, deep-chested roar that transitions into a high-pitched wail.
TAKE_C: Strained, breathless scream that peters out into a gasp at the very end.
Prosody: Sudden onset at 00:03, maximum volume at 00:05, slight taper at 00:07.
Sync: Mouth must be at maximum aperture during the peak of the audio (00:04-00:06).
Mic Signature: Close-mic'ed, high saturation, slight distortion to indicate power, dry room tone.

How soy_aria_cruz Made This Angry Scream AI Video — and How to Recreate It

This case study analyzes a high-intensity AI video generation test by creator @soy_aria_cruz, showcasing the emotional range of the Seedance 1.5 Pro model. The video features a "Cinematic Angelic Portrait" that subverts expectations: instead of a serene celestial being, the subject delivers a raw, visceral "Angry Scream" (Grito Enfadado). This specific piece of content leverages the "AI Stress Test" trend, where creators push video models to their limits to see if they can maintain character consistency during extreme facial deformations. With its high-contrast lighting, glowing halo, and shocking emotional shift, it serves as a perfect template for creators looking to showcase technical AI mastery while maintaining high aesthetic standards.

What You’re Seeing

The video presents a medium close-up of a young woman with dark hair tied back, wearing large round glasses and an abundance of layered gold and pearl necklaces. She is dressed in a gold sequined camisole, but the most striking elements are the large, realistic white feathered wings and a glowing orange neon halo floating above her head. The background is a dark, moody environment with soft bokeh, making the subject and her glowing accessories pop. The color grade is warm and cinematic, with a focus on skin texture and the reflective surfaces of her jewelry.

Shot-by-Shot Breakdown

Time Range	Visual Content	Shot Language	Lighting & Tone	Viewer Intent
00:00–00:03	Subject looks at the camera with a frustrated, simmering anger.	Medium Close-up (MCU), static.	Warm halo rim light, cool shadows.	The Hook: Establish tension and character.
00:03–00:08	Subject opens her mouth in a full-throated, visceral scream.	MCU with slight camera shake.	Flickering light from the halo.	The Payoff: Demonstrate AI's ability to handle extreme motion.

Why It Went Viral

The success of this video lies in the "AI Capability Reveal." For the current audience of AI enthusiasts and creators, the "holy grail" isn't just a pretty image; it's temporal consistency during high-motion tasks. Screaming is notoriously difficult for AI to render without the face "melting" or losing the character's identity. By choosing an "Angelic" subject—traditionally associated with peace—and forcing her into a state of "Rage," the creator creates a psychological dissonance that demands attention.

From a platform perspective, this video triggers high "Save" and "Share" rates. Other creators save it as a benchmark for what Seedance 1.5 Pro can achieve, while casual viewers share it because of the "uncanny valley" fascination—it looks so real yet is clearly artificial. The use of clear text overlays ("SEEDANCE 1.5 PRO") also positions the content as a "tech review," which attracts a specific, high-engagement niche interested in the latest tools.

5 Testable Viral Hypotheses

The Contrast Hook: If you pair a "pure" character (angel, nun, child) with an "extreme" emotion (rage, terror), watch time increases due to cognitive dissonance.
The Tool Benchmark: Explicitly naming the AI model in the video (e.g., "Sora," "Kling," "Seedance") attracts the "Early Adopter" crowd who will comment to discuss the tech.
The "Stress Test" Format: Showing a character performing a difficult physical action (eating, screaming, crying) proves the AI's quality, leading to higher "Save" counts for reference.
The Glow Aesthetic: Using high-contrast "rim lighting" (like the halo) masks AI artifacts and creates a high-end cinematic feel that stops the scroll.
The Audio-Visual Sync: A loud, jarring audio hook (the scream) synchronized with the visual peak forces the viewer to stop scrolling and pay attention to the "why."

How to Recreate (Step-by-Step)

1. Topic Selection & Positioning

This video suits "AI Art" or "Tech Review" accounts. Choose a character archetype (e.g., Cyberpunk Samurai, Victorian Ghost) and an extreme emotion that contradicts that archetype.

2. Character Consistency

Generate a high-quality base image using Midjourney or DALL-E 3. Use a prompt that specifies unique identifiers: "Round gold glasses," "layered pearl necklaces," "ponytail." This ensures the AI has "anchors" to hold onto during the video generation.

3. Define the "Glow"

Include a "practical light source" in your prompt, like a neon halo or a glowing orb. This provides a logical reason for the cinematic lighting and helps the AI maintain consistent shadows on the face.

4. Video Generation (The Scream)

Upload your base image to an Image-to-Video tool (like Seedance, Kling, or Luma). Use a motion prompt: "The woman looks at the camera with frustration, then opens her mouth wide and lets out a powerful, visceral scream, facial muscles tensing, realistic skin deformation."

5. Adjusting Motion Strength

If the face "breaks," lower the motion strength. If the scream is too subtle, increase it. The goal is to find the "sweet spot" where the mouth opens fully but the glasses stay on the face.

6. Adding the Audio Hook

The audio is 50% of the impact. Use a high-quality foley sound of a scream. Sync the peak of the audio exactly with the frame where the mouth is widest.

7. Title & Branding

Add clean, minimalist text overlays. Mention the tool used. This builds authority and makes the content searchable for people looking for that specific AI model.

8. Platform Adaptation

For Instagram/TikTok, use a "loop" effect. Ensure the first frame is the most visually striking (the glowing halo) to act as a thumbnail.

Growth Playbook

Opening Hook Lines

"Can AI actually feel rage? Testing Seedance 1.5 Pro..."
"The most realistic AI scream I've seen yet. 🤯"
"Stop scrolling: This is the future of AI emotional range."

Caption Templates

The Tech Reviewer:
Testing the emotional limits of [AI Model Name] today. 🤯 I used the same prompt for 5 different models, but this one handled the 'Angry Scream' best. What do you think of the facial consistency? 👇
#AI #VideoAI #Seedance #CreativeTech

The Storyteller:
Even angels have bad days. 😇🔥 Pushing the boundaries of AI storytelling with this one. The detail in the facial muscles is insane. Should I do a 'Crying' version next?
#AIFilm #CinematicAI #DigitalArt #Emotion

Hashtag Strategy

Broad: #AI #ArtificialIntelligence #DigitalArt #VideoProduction
Mid-tier: #AIVideo #GenerativeAI #AIArtist #Cinematic
Niche: #Seedance #AIModelTest #TemporalConsistency #AIFilmmaking

FAQ

What tools make it look the most similar?

Seedance 1.5 Pro, Kling AI, or Luma Dream Machine are best for high-intensity facial expressions.

What are the 3 most important words in the prompt?

"Visceral," "Facial deformation," and "Consistency."

Why does the generated face look inconsistent?

Usually, the motion strength is too high or the base image lacks enough "anchor points" like glasses or jewelry.

How can I avoid making it look like AI?

Add film grain, a slight camera shake, and ensure the lighting is "motivated" by an object in the scene.

Is it easier to go viral on Instagram or TikTok with this?

Instagram Reels currently favor high-aesthetic "Cinematic AI," while TikTok favors "Tutorial/Process" AI content.