> For the complete documentation index, see [llms.txt](https://docs.qolaba.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.qolaba.ai/video-generation/output-controls/aspect-ratio-dimensions-and-audio.md).

# Aspect Ratio, Dimensions & Audio

Aspect ratio defines the shape and framing of your video. Audio generation adds an AI-generated sound layer to your output. Both are configured in the Output Controls before generating.

***

#### Aspect Ratio & Dimensions

Aspect ratio has no effect on credit cost but significantly impacts how the video performs across platforms and how well it uses any reference images uploaded as inputs.

**Available Aspect Ratios**

| Aspect Ratio | Orientation | Best For                                         |
| ------------ | ----------- | ------------------------------------------------ |
| **16:9**     | Landscape   | YouTube, presentations, desktop, TV, widescreen  |
| **9:16**     | Portrait    | Instagram Reels, TikTok, YouTube Shorts, Stories |
| **1:1**      | Square      | Instagram feed, LinkedIn, social media posts     |
| **4:3**      | Landscape   | Classic film look, older display formats         |
| **3:4**      | Portrait    | Pinterest, some mobile formats                   |
| **21:9**     | Ultrawide   | Cinematic widescreen, film production            |

**Platform Quick Reference**

| Platform        | Recommended Ratio |
| --------------- | ----------------- |
| YouTube         | 16:9              |
| Instagram Feed  | 1:1               |
| Instagram Reels | 9:16              |
| TikTok          | 9:16              |
| YouTube Shorts  | 9:16              |
| Pinterest       | 3:4               |
| LinkedIn        | 16:9 or 1:1       |
| Cinema / Film   | 21:9              |

**Matching Aspect Ratio to Reference Images**

When uploading an initial image or last frame image, match the output aspect ratio to the source image's natural dimensions. Mismatched ratios result in either cropped or stretched framing in the generated output.

{% hint style="info" %}
Before uploading a reference image, check its dimensions and select the closest matching aspect ratio. A portrait photo (3:4) used with a 16:9 output will be significantly cropped and may exclude key subject elements. See [Reference Media Uploads →](/video-generation/input-controls/reference-media-uploads.md) for more guidance.
{% endhint %}

***

#### Audio Generation Controls

Audio generation is available exclusively on **Seedance** and **Seedance 2.5**. All other models generate silent video only.

When enabled, Qolaba generates a contextually matched audio layer alongside the video — ambient sound, environmental audio, or scene-appropriate sound based on the visual content.

**Audio On vs. Audio Off**

| Setting                | Output                        | Credit Impact  | When to Use                                              |
| ---------------------- | ----------------------------- | -------------- | -------------------------------------------------------- |
| **Audio ON** (default) | Video with AI-generated audio | Standard       | Final outputs for social media, demos, storytelling      |
| **Audio OFF**          | Silent video                  | Slightly lower | When adding custom audio in post, or audio is not needed |

**How to Toggle Audio**

1. Select **Seedance** or **Seedance 2.5** as your model
2. The **Audio Generation toggle** becomes available in Output Controls
3. Toggle **on** to include audio or **off** to generate silent video
4. The credit counter updates to reflect the change

{% hint style="info" %}
The Audio toggle is currently only available with Seedance model.
{% endhint %}

**When to Disable Audio**

* You plan to add your own music, voiceover, or sound effects in post-production
* The video will play in a muted or silent-autoplay context
* You want to reduce credit usage slightly during draft generations
* The AI-generated audio from a previous run did not match the intended tone

**Using Audio with Other Models**

If you need audio alongside video from a non-Seedance model, generate the video first using your preferred model, then use the [Text-to-Speech tool →](/speech-generation/introduction.md) to create a separate audio layer and combine both in post-production.