# How to Make the Viral Hotel Lobby AI Video With 2 Photos (2026 Guide + Prompts)

Learn how to make the viral Hotel Lobby AI video from two photos. Explore free options, copy-paste prompts, audio tips, and fixes for common AI video errors.

Canonical URL: https://aiidelist.com/blog/how-to-generate-hotel-lobby-ai-video

Language: en

Published: 2026-10-16

Updated: 2026-10-16

## Key Takeaways

- **The Hotel Lobby AI video trend** recreates the distinctive orange-studio rap performance associated with Quavo and Takeoff's 2022 appearance on A COLORS SHOW, replacing the performers with friends, couples, pets, or fictional characters.
- **The easiest method requires two photos.** Upload one clear portrait for each performer to a compatible Hotel Lobby AI video generator, select the desired format, and generate the duet.
- **No dedicated generator is required.** General AI image and video tools can create an original orange-booth performance using reference images and carefully structured prompts.
- **A convincing result depends on five elements:** a consistent orange background, one hanging microphone, two recognizable performers, stable left-right positioning, and believable synchronized movement.
- **Free options exist, but limitations vary.** Some services offer introductory credits or previews, while longer videos, higher resolutions, or watermark-free exports may require payment.
- **Music and disclosure matter.** Use appropriately licensed audio, obtain permission for recognizable likenesses, and label realistic AI-generated content when publishing.

## What Is the Hotel Lobby AI Video Trend?

The Hotel Lobby AI trend is a viral short-video format in which AI-generated performers appear inside a minimalist orange recording studio, rapping and gesturing around a single microphone suspended from the ceiling.

Despite its name, the trend does not take place inside a hotel lobby. It draws inspiration from [Quavo and Takeoff's performance of HOTEL LOBBY on A COLORS SHOW](https://www.youtube.com/watch?v=x9yop0nYR9g), published on June 17, 2022.

The performance features the two artists, also known as Unc & Phew, sharing a microphone against a striking monochromatic background. Its simple composition became particularly recognizable when AI-generated reinterpretations spread across TikTok, Instagram Reels, and other social platforms in September 2026.

The resurgence was significant enough that Quavo discussed the phenomenon in an [October 2026 Associated Press interview](https://apnews.com/article/92c67c663ef789c7df8310ff09b3406b), reflecting on both the renewed popularity of the song and the emotional connection to Takeoff, who died in 2022.

The familiar AI format follows a consistent structure:

- **Background:** A seamless, saturated burnt-orange studio backdrop.
- **Performers:** Two recognizable subjects standing side by side.
- **Microphone:** One centrally positioned microphone hanging from above.
- **Camera:** A mostly stationary, front-facing shot.
- **Movement:** Rhythmic head movements, hand gestures, and alternating performances.
- **Duration:** Usually a short clip suitable for TikTok, Reels, or YouTube Shorts.

The format works because viewers immediately recognize the scene, even when the performers are replaced with unexpected characters.

A video featuring two cats, for example, becomes recognizable through the same orange background and shared microphone without requiring additional explanation.

## How to Generate a Hotel Lobby AI Video: The Quickest Method

For beginners, a dedicated Hotel Lobby AI generator is usually the simplest option. These services automate the background, performer placement, and much of the animation process.

**Step 1: Prepare two photos**

Choose one image for each performer. Front-facing portraits with clear facial features and even lighting provide the strongest starting point.

**Step 2: Open a compatible generator**

Dedicated services such as [HotelLobby.me](https://hotellobby.me/how-to-make-hotel-lobby-ai-video) and template-based tools such as [Picsman AI](https://www.picsman.ai/blog/how-to-make-hotel-lobby-ai-video/) provide examples of photo-to-video workflows for this trend.

Compare their current preview options, prices, output quality, privacy policies, and audio handling before uploading personal photos.

**Step 3: Upload both images**

Assign the first photo to the left performer and the second to the right performer, following the selected tool's interface.

Choose the performer order carefully. Some generators cannot change identities or positions after rendering.

**Step 4: Select video settings**

When supported, choose:

- Aspect ratio: **9:16** for TikTok, Reels, and Shorts.
- Duration: Approximately **10–15 seconds** for a compact performance.
- Resolution: The highest practical setting, preferably 1080p where available.
- Style: Realistic studio performance or the closest available orange-booth preset.

Individual generators may offer only fixed durations, resolutions, or aspect ratios.

**Step 5: Generate and review**

Start the generation and inspect the completed clip before downloading.

Pay particular attention to facial consistency, distorted fingers, performer positions, microphone placement, and whether the video includes audio.

**Step 6: Export and share**

Download the MP4 file. If the generator outputs a silent video, add suitable audio in a video editor or through the publishing platform's music library.

The result is a short AI duet built from two reference photos without filming a live performance.

## How to Make a Hotel Lobby AI Video Using AI Prompts

Creators seeking more control over facial identity, costumes, camera composition, or choreography can use a general-purpose AI image and video workflow.

This method is more flexible than a fixed template, but it also requires more careful preparation.

A particularly reliable approach is to separate the process into two stages:

1. Generate a high-quality still image showing both performers in the desired environment.
2. Animate that completed image using an image-to-video model.

This reduces the number of things the video model must invent simultaneously.

For example, [Runway's reference-image documentation](https://help.runwayml.com/hc/en-us/articles/40042718905875-Creating-with-Gen-4-Image-References) describes how multiple visual references can guide image generation and help preserve recognizable subjects across scenes.

The resulting composition can then serve as the starting frame for a compatible video model.

## Step 1: Prepare Reference Photos for Better Face Consistency

Image quality strongly influences whether the generated performers remain recognizable throughout the animation.

Use the following guidelines:

| Photo requirement | Recommended approach | Why it matters |
|---|---|---|
| Face angle | Front-facing or slight three-quarter view | Reduces identity ambiguity |
| Lighting | Soft, even illumination | Makes facial structure easier to preserve |
| Resolution | Clear, detailed original photos | Reduces artifacts introduced by low-quality references |
| Framing | Head-and-shoulders or waist-up | Provides facial and clothing details |
| Background | Simple, uncluttered | Helps separate the subject from the environment |
| Number of people | One person per reference | Reduces unintended identity mixing |

Avoid extreme beauty filters, heavily compressed screenshots, faces obscured by sunglasses, and photos containing multiple competing subjects.

For a person-and-pet duet, provide a clear photograph of each subject separately.

If the generator supports only one starting image, first combine both subjects into a single AI-generated composition rather than expecting the animation stage to reconstruct two identities from unrelated photos.

## Step 2: Generate the Orange-Studio Starting Image

Use an image generator that supports reference images to produce the initial scene.

**Copy-and-paste Hotel Lobby AI image prompt:**

```text
Create a photorealistic music-performance scene using the TWO provided reference images.

SUBJECT A: Preserve the identity, facial structure, hairstyle, skin tone, and recognizable clothing of the person in reference image 1. Position Subject A on the LEFT.

SUBJECT B: Preserve the identity, facial structure, hairstyle, skin tone, and recognizable clothing of the person in reference image 2. Position Subject B on the RIGHT.

SCENE: Both performers stand inside a minimalist recording studio with a seamless matte burnt-orange background and matching floor.

A single black professional microphone hangs vertically from the ceiling, positioned precisely between the two performers.

COMPOSITION: Static front-facing camera, centered medium-wide composition, both performers clearly visible, realistic body proportions, adequate spacing, and balanced studio lighting.

VISUAL STYLE: Photorealistic, cinematic studio photography, natural skin texture, subtle shadows, realistic clothing details, warm color palette, high-definition detail.

CONSTRAINTS: Exactly two performers, one hanging microphone, consistent facial identity, no extra people, no additional microphones, no captions, no watermarks, no logos.

OUTPUT: Vertical 9:16 composition suitable for a short-form music-performance video.
```

The most important constraints are the performer positions, central microphone, and stable camera framing.

A common mistake is requesting excessive cinematic movement before establishing a consistent composition. Start with the simplest possible scene and introduce additional complexity only after the characters look correct.

## Step 3: Animate the Image Into a Hotel Lobby AI Video

Once the still image looks convincing, upload it to an image-to-video model.

Use a motion-focused prompt instead of repeating the entire visual description.

**Copy-and-paste Hotel Lobby AI video prompt:**

```text
Animate the provided reference image into an original 12-second photorealistic rap-duet performance.

Maintain the exact studio layout, orange background, clothing, facial identities, and central hanging microphone from the reference image.

Subject A remains on the LEFT throughout the clip.
Subject B remains on the RIGHT throughout the clip.

ACTION SEQUENCE:
0–4 seconds: Subject A performs confidently toward the camera with subtle rhythmic head movement and controlled hand gestures. Subject B listens, nods, and reacts naturally.

4–8 seconds: Subject B takes the lead with expressive but realistic hand movements. Subject A responds with subtle nodding and a relaxed posture.

8–12 seconds: Both performers briefly acknowledge one another, then face forward and continue moving naturally to the imagined beat.

CAMERA: One continuous locked-off front-facing shot. No zooming, camera rotation, or scene transitions.

MOTION: Natural body movement, realistic hands, consistent facial structure, believable timing, subtle expressions, and independent performer reactions.

Keep the scene visually stable. Preserve the same two identities in every frame.

Do not add extra performers, microphones, objects, text, logos, or unexpected scene changes.
```

The timing instructions provide a useful target rather than guaranteed frame-accurate choreography. Different video models interpret motion and duration constraints differently.

**Recommended starting settings:**

| Setting | Suggested value |
|---|---|
| Aspect ratio | 9:16 |
| Initial preview | 4–6 seconds when supported |
| Final duration | 10–15 seconds |
| Camera motion | Static |
| Movement complexity | Low to moderate |
| Output format | MP4 |
| Export target | 1080 × 1920 when available |

A short preview can reveal whether the model preserves both identities before credits are spent on a longer render.

## Step 4: Use a Negative Prompt to Reduce AI Artifacts

Some video generators provide a separate negative-prompt field. Others respond better when important constraints are expressed directly in the main prompt.

When negative prompting is supported, try:

```text
extra people, duplicate faces, merged identities, swapped performers, distorted fingers, extra limbs, missing hands, inconsistent clothing, unnatural face morphing, duplicate microphones, floating objects, camera shake, sudden zoom, scene cuts, blurry facial features, inconsistent lighting, flickering background, text, logos, watermarks
```

Negative prompts are not guaranteed to eliminate these problems.

The more reliable strategy is to reduce unnecessary motion, strengthen the reference image, and correct one specific failure at a time.

## Alternative Method: Create the Trend With Motion Transfer

The prompt-based workflow generates new choreography. A motion-transfer workflow instead uses an authorized reference performance to guide movement.

This distinction matters because many recognizable Hotel Lobby AI videos imitate a particular sequence of gestures rather than simply showing two performers rapping.

A typical motion-transfer workflow involves:

1. Preparing an authorized reference clip containing the intended movements.
2. Supplying reference images for the replacement performers.
3. Mapping the intended subjects to the correct positions.
4. Generating the replacement motion.
5. Compositing or refining the results if the model cannot animate both subjects independently.

**Important technical limitation:** Some motion-control systems are optimized for one primary subject. When a reference video contains two equally prominent performers, the system may animate only one person or confuse their movements.

For these systems, a more advanced approach is to isolate the movements of each performer and generate them separately before compositing the final duet.

This requires additional editing and is not the easiest route for beginners.

For an original creation, generating fresh choreography is generally simpler than attempting a frame-for-frame reconstruction of copyrighted performance footage.

## Which Hotel Lobby AI Video Method Is Best?

The right workflow depends on the desired balance of speed, customization, and technical control.

| Method | Best for | Main advantage | Main limitation |
|---|---|---|---|
| Dedicated Hotel Lobby AI generator | Beginners | Minimal setup, usually no prompt required | Limited control over choreography and styling |
| Template-based AI video tool | Social-media creators | Fast and convenient | Template availability and restrictions vary |
| AI image + image-to-video | Custom scenes and characters | Greater control over composition and appearance | Additional prompting and retries |
| Motion-transfer workflow | Matching a specific performance | More direct motion guidance | Multi-person tracking and rights complexity |
| Manual video editing | Experienced creators | Precise control over timing and finishing | More production effort |

**Practical recommendation:** Start with a specialized two-photo generator when the objective is simply to participate in the trend. Choose an image-to-video workflow when identity accuracy, original styling, or creative flexibility matters more than convenience.

## Can You Generate a Hotel Lobby AI Video for Free?

Potentially, yes. However, a free preview is not necessarily equivalent to a complete, downloadable video.

Free access varies by provider and may involve:

- A limited number of introductory credits.
- Shorter preview durations.
- Lower-resolution renders.
- Watermarked downloads.
- Restricted use of premium templates.
- Additional credits for longer videos or multiple generations.

Services change their pricing and credit policies frequently. Check the current generation cost, whether failed renders are refunded, and whether the free tier supports MP4 downloads before starting.

For a more economical workflow, prepare the composition carefully and generate a short preview before purchasing a longer export.

## Best Hotel Lobby AI Prompt Ideas for Different Characters

The original appeal of the trend comes from replacing the expected performers with an unexpected pairing.

Once a basic workflow is working, the same composition can support multiple variations.

**Friends or couples:**

```text
Two consenting adult friends perform a playful rap duet in a minimalist burnt-orange recording studio. One performer stands on the left, the other on the right, with a single hanging microphone between them. Both retain the facial features of their supplied reference photos. They exchange confident gestures and rhythmic reactions while remaining in their original positions. Static camera, natural movement, cinematic lighting, vertical 9:16 video.
```

**Person and pet:**

```text
A photorealistic adult and their pet dog perform a humorous music-video duet inside a seamless orange studio. The adult stands on the left and the recognizable dog from the reference photo stands on the right. A single microphone hangs between them. The adult performs subtle rhythmic gestures while the dog reacts with gentle head tilts and natural movements. Preserve the dog's markings and appearance. Static front-facing camera, playful but believable motion, no extra subjects, vertical 9:16.
```

**Animated characters:**

```text
Two original stylized animated characters perform a playful rap duet in a bright orange recording studio. One character remains on the left and the other on the right, with a single suspended microphone in the center. Preserve the characters' distinct costumes, proportions, and facial designs. Add expressive hand gestures, rhythmic movements, and alternating reactions. High-quality 3D animation, stable composition, locked camera, vertical 9:16.
```

The prompts should be adapted to the selected model's supported inputs. A textual prompt alone cannot guarantee the accurate preservation of two reference identities.

## How to Add the Hotel Lobby Sound to Your AI Video

Audio is one of the most important details separating a generic rap animation from a recognizable Hotel Lobby-style video.

The trend is associated with Quavo and Takeoff's *Hotel Lobby (Unc & Phew)* performance, but not every generator includes that soundtrack.

Some tools generate a silent video, some create original AI audio, and others may include music within the exported file.

For a silent export intended for TikTok:

1. Open TikTok and upload the generated video.
2. Select the sound option in the video editor.
3. Search for the relevant Hotel Lobby audio or open an existing trend video and select its available sound.
4. Adjust the audio starting point to match the gestures.
5. Preview the complete video before publishing.

The same basic process applies to Instagram Reels using its available music-selection tools.

**Copyright consideration:** Music availability varies by region, account type, and licensing restrictions. The presence of audio in a social platform's library does not automatically authorize commercial advertising, redistribution, or use on another platform.

For commercial projects, original or appropriately licensed music may be the more suitable choice.

For precise synchronization, plan gestures around musical accents before rendering or edit the final clip against the approved audio afterward. Generating video and audio separately often provides more control than relying on automatic lip synchronization.

## Why Your Hotel Lobby AI Video Looks Wrong: Common Problems and Fixes

Even advanced AI video models can struggle with two recognizable subjects interacting in the same frame.

| Problem | Likely cause | Recommended fix |
|---|---|---|
| Faces change during movement | Weak reference identity or excessive motion | Use clearer portraits and reduce head rotation |
| Performers swap positions | Ambiguous left-right instructions | Explicitly assign and lock both positions |
| Only one performer moves | Single-subject animation limitation | Use a two-person-capable workflow or separate compositing |
| Extra fingers or distorted hands | Complex gestures and occlusion | Request smaller, slower hand movements |
| Microphone changes shape | Unstable object generation | Use a fixed starting frame and minimal camera movement |
| Mouth movements do not match music | Audio and video generated independently | Adjust audio timing or use supported lip-sync editing |
| Background changes color | Weak environment consistency | Reuse a fixed reference image and simplify the prompt |
| Faces look blurry | Poor source images or low output resolution | Replace the references and select a higher-quality render |
| Subjects are cropped | Incorrect framing or aspect ratio | Regenerate in 9:16 with both subjects inside safe margins |

**A useful debugging rule is to change one variable per attempt.**

If facial identity is unstable, improve the reference photographs before changing the music, camera, background, and choreography simultaneously.

For challenging two-person scenes, compare three checkpoints: the opening frame, a middle frame, and the final frame. A result that looks convincing at the start can still contain substantial identity drift by the end.

## Advanced Tips for More Realistic Hotel Lobby AI Videos

**1. Build the scene before animating it.** A stable, recognizable starting frame gives the video model less room to reinterpret the environment and performer identities.

**2. Keep the camera stationary.** Static framing reduces the need to reconstruct the scene from multiple viewpoints and better matches the recognizable orange-studio presentation.

**3. Give the performers separate actions.** Two people moving identically can look synthetic. Alternating lead and reaction behavior creates a more believable performance.

**4. Keep gestures manageable.** Fast hand movements near faces increase occlusion and can produce distorted anatomy or identity changes.

**5. Separate creative iteration from final rendering.** Test the composition with short, lower-cost previews, then use higher-quality output settings after identifying a reliable prompt.

**6. Check the entire video at normal playback speed.** Inspect not only image quality but also timing, flicker, unnatural body motion, and abrupt transitions.

**7. Avoid unnecessary upscaling.** Increasing resolution cannot reliably restore facial details that were never generated correctly. Fix identity and motion problems before applying finishing enhancements.

These techniques are useful beyond the Hotel Lobby trend and apply to other multi-character AI music videos, cinematic animations, and social-media short-form content.

## How to Export the Video for TikTok, Reels, and YouTube Shorts

For a typical vertical short-form video, the following export settings provide a practical target:

| Setting | Recommended export |
|---|---|
| Orientation | Vertical |
| Aspect ratio | 9:16 |
| Resolution | 1080 × 1920 if available |
| Container | MP4 |
| Video codec | H.264 for broad compatibility |
| Frame rate | 24 or 30 fps, matching the source where practical |
| Audio | Appropriately licensed soundtrack |

Keep both faces and the microphone away from the extreme edges of the frame, where social-platform interface elements may obscure important details.

For horizontal versions, use a separately generated 16:9 composition when possible. Simply cropping or stretching an existing vertical clip can damage the framing.

A short caption can communicate the creative pairing without explaining the entire trend. For example:

- Two best friends. One mic. Zero preparation.
- The rap duo nobody expected.
- Our pets finally started a music career.
- When the group chat becomes a recording studio.

Captions should describe the actual characters in the clip rather than inventing events or suggesting the performance was genuinely filmed.

## Copyright, Consent, and AI Content Labels

The Hotel Lobby AI trend raises questions about creative ownership and digital likeness, especially because it is inspired by a recognizable performance involving a deceased artist.

Creating an original scene inspired by the orange-studio format is not the same as obtaining permission to reuse copyrighted footage, music, or another person's likeness.

Creators should consider four issues before publication:

- **Consent:** Use photographs of recognizable private individuals only with appropriate permission. Avoid creating misleading impersonations.
- **Copyright:** Check the rights associated with source footage, music, choreography, and other protected material. Permission to use one asset does not grant rights to the others.
- **Privacy:** Review the generator's terms, retention policy, and treatment of uploaded photographs, particularly for sensitive personal images.
- **Transparency:** Clearly disclose realistic AI-generated performances that could be mistaken for real recordings.

[TikTok's AI-generated content guidance](https://support.tiktok.com/en/using-tiktok/creating-videos/ai-generated-content) requires creators to label realistic AI-generated content and restricts certain uses of real people's likenesses, including non-consensual depictions of private individuals.

[YouTube's generative-AI disclosure policy](https://support.google.com/youtube/answer/14328491) likewise requires disclosure when realistic synthetic content depicts events or actions that did not actually occur.

An AI label is a transparency measure, not a substitute for permission or licensing. Requirements may also differ between personal entertainment and commercial distribution.

## Frequently Asked Questions About Hotel Lobby AI Videos

**How do you make a Hotel Lobby AI video from two photos?**

Upload one clear reference photo for each performer to a compatible Hotel Lobby AI generator. Assign their positions, choose the available output settings, generate the clip, and review it before downloading. Alternatively, create a composite image of both people and animate it using an image-to-video model.

**What is the Hotel Lobby AI video trend based on?**

It is inspired by Quavo and Takeoff's 2022 performance of *Hotel Lobby (Unc & Phew)* on A COLORS SHOW, featuring the artists in an orange studio with a suspended microphone.

**Is there a free Hotel Lobby AI video generator?**

Some AI video services provide introductory credits, free trials, or short previews. Full-length, high-resolution, watermark-free generation may require payment. Availability and restrictions depend on the provider.

**Can you create a Hotel Lobby AI video without a special app?**

Yes. An AI image generator and a compatible image-to-video model can create an original version of the scene. This approach offers more control but may require additional editing.

**Can you make the Hotel Lobby AI video with pets?**

Yes. Cats, dogs, and original fictional characters can be used when the chosen generator supports their appearance and movement. Results vary, particularly when the model attempts to animate animals performing human-like gestures.

**Does the video need the original Hotel Lobby song?**

No. The orange-studio visual format can work with an original or appropriately licensed soundtrack. Using the familiar song makes the reference more recognizable, but music rights and platform restrictions still apply.

**Why do both characters sometimes turn into the same person?**

This is a multi-subject identity-consistency problem. Using separate clear references, fixed left-right assignments, and a single composed starting image can help reduce identity blending.

**Can one group photo be used instead of two individual photos?**

That depends on the generator. Some tools accept group photographs or support subject selection, while others require two separate portraits. For the most predictable result, prepare individual reference images.

**How long should a Hotel Lobby AI video be?**

Approximately 10–15 seconds is a useful target for a short performance. Individual generators may impose different fixed durations or allow shorter and longer outputs.

**Do you need CapCut to create the Hotel Lobby trend?**

No. CapCut is an optional editing and template tool. The animation can be generated elsewhere and uploaded directly to a social platform, although editing software can help refine timing, captions, cropping, and sound.

## Conclusion

Generating a Hotel Lobby AI video is now accessible to creators without professional video-editing experience. The simplest workflow uses two clear photos and a specialized AI generator to recreate the instantly recognizable orange-studio duet format.

For creators who want greater control, combining reference-based image generation with image-to-video animation offers more flexibility over character identity, camera framing, performance style, and choreography.

The most effective approach is to prioritize **accurate reference images, stable performer placement, simple camera movement, and careful audio synchronization** rather than relying on increasingly complicated prompts.

Start with two authorized photos, test a short generation, and refine the result before exporting. With the right composition and a distinctive pairing, the Hotel Lobby-inspired format can become an entertaining, original AI video suitable for TikTok, Instagram Reels, or YouTube Shorts.