Video to Reel Maker
Turn a horizontal video into a vertical 9:16 reel with a Pro style timeline trimmer and automatic captions. Crop to fill means no blur bars, only real footage. Drag inside the preview to pick the crop. Everything runs in your browser.
100% secure. Your video never leaves your device.
Related tools
Everything runs in your browser. No uploads, no sign-up.
Video Trimmer
Cut start and end without quality loss.
Video Compressor
Shrink video for uploads and messages.
Video to GIF
Turn a clip into an animated GIF.
Video to MP3
Extract audio from any video.
Video Upscaler
Enlarge video 2x, 3x, or 4x in your browser.
Audio Recorder
Record and trim audio with a live waveform.
What is a video to reel maker
A video to reel maker converts a traditional horizontal video, shot in a 16:9 landscape frame, into a vertical 9:16 clip that fills the entire phone screen. Reels, Shorts, and TikTok all use the vertical format, and horizontal uploads either show up with black bars on the sides or get cropped awkwardly by the platform. This tool gives you control over how the conversion happens.
Instead of uploading the file to a server, everything happens in your browser using two modern web APIs: WebCodecs for hardware accelerated encoding, and WebGPU for running the Whisper speech recognition model on your GPU. There is no account, no queue, no watermark, and no daily limit. You can use it as often as you want, offline, without giving anyone your video.
The timeline works like a professional editor. You drag the teal handles inward to set the start and end of the clip. The selected range appears as thumbnails you can visually scan, and only that range is rendered to keep exports fast. Inside the preview you can also drag to pan the crop, choosing which part of the source footage fills the vertical frame. Generate captions and they burn directly into the video, in sync with the audio down to the word.
What it does
Auto captions with Whisper
On-device speech recognition with word level accuracy. Model runs on your GPU through WebGPU. No API keys, no server, no cost.
Premiere style timeline trim
Drag the teal handles on the timeline to set the start and end of your clip, just like a desktop editor. Only the highlighted range renders.
Drag to pick the crop
Drag inside the preview to pan the video and choose which part stays visible. Pick the left, right, top, or bottom of the frame with one drag.
Crop to fill, no blur bars
The video is zoomed and cropped to fill the target frame. No blurred side bars. Every pixel is real footage.
Four social formats
9:16 for Reels, Shorts, and TikTok. 3:4 and 4:5 for Instagram feed. 1:1 for square posts. Switch instantly and the preview updates.
Hardware accelerated encoding
Uses WebCodecs, the browser native H.264 encoder. Up to 20 times faster than ffmpeg.wasm based tools, with smaller output files.
Customizable captions
Pick top, middle, or bottom. Adjust the font size from 36 to 80 pixels. The live preview matches the final render exactly.
Works offline after first load
Once the models are cached by the browser, everything runs without a network connection. Your video never touches a server.
No sign-up, no watermark
No account, no email, no paywall, no overlay on the output. Open the page, drop a video, download the reel.
How it works
Step 1
Drop your video
Drag any horizontal or vertical video onto the page. MP4, WebM, MOV, and MKV are all supported. Nothing is uploaded.
Step 2
Trim and pick the crop
Drag the teal handles on the timeline to select the exact range. Drag inside the preview to choose which part of the video stays visible.
Step 3
Render and download
Click Render. WebCodecs encodes the reel on your device. Download the MP4 and upload it to any platform.
Common use cases
The tool fits any workflow that needs to convert horizontal footage into a social ready vertical clip. Here are the most common ones.
Repurpose YouTube videos
Turn a long horizontal YouTube video into a short vertical reel for TikTok, Instagram, or YouTube Shorts. Just drop the video, trim to the best 30 seconds, and generate captions automatically.
Repurpose landscape vlogs
Vlogs shot in 16:9 become vertical shorts with a single trim. Captions built for silent viewing make them perform on mobile feeds.
Convert product demos
Record a horizontal product demo and cut it into a vertical teaser. The crop to fill keeps the product in frame without awkward black bars.
Podcast clips
Cut a 60 second highlight from a long podcast and add burned in captions so the clip works with sound off.
Course previews
Convert a horizontal lesson into a vertical preview clip. Use the timeline to pick the most engaging segment and set the aspect for LinkedIn or Instagram.
Real estate and travel
Turn landscape property or travel footage into vertical reels that fill the whole phone screen. Crop to fill means no letterboxing.
Best practices for reels and shorts
A few habits make the difference between a reel that gets scrolled past and one that gets watched to the end. These tips come from how the major platforms rank short form video.
Aim for 15 to 30 seconds
Reels and Shorts that perform best on platforms are usually 15 to 30 seconds. Use the Trim to 15s and Trim to 30s preset buttons to stay in the sweet spot.
Lead with the hook
Cut off the intro and start at the most interesting moment. The timeline lets you skip straight to it, so the first second captures attention.
Always add captions
A very large share of feed viewers watch with sound off. Captions are the single biggest lever you have on retention and completion rate.
Keep the safe zone in mind
Instagram and TikTok overlay UI at the bottom of the frame. Use the caption position set to Middle or Top if your title overlaps the buttons.
Match the platform ratio
Reels, Shorts, and TikTok are 9:16. Instagram feed posts are 4:5. LinkedIn and Facebook square posts are 1:1. Pick the right one at export.
Check the first frame
The first frame shows as the cover on some platforms. Trim so the first frame is not mid blink or motion blur.
How it compares
Most online reel makers are cloud services: you upload your video, wait in a queue, and get a link back. That model works, but it means your footage leaves your computer. ToolDeck takes a different approach.
| What matters | Typical tools | ToolDeck |
|---|---|---|
| Works offline | No | Yes, after first load |
| Sign-up required | Usually | No |
| File upload | Yes, to a server | Never, runs in your browser |
| Watermark | Common on free plans | Never |
| Export length limit | Common on free tiers | None, bounded by device memory |
| Auto captions | Often paid | Free, on-device |
| Cost | Credits or subscription | Free, unlimited |
Related tools
Video Trimmer
Cut start and end without quality loss.
Video Compressor
Shrink video for uploads and messages.
Video to GIF
Turn a clip into an animated GIF.
Video to MP3
Extract audio from any video.
Video Upscaler
Enlarge video 2x, 3x, or 4x in your browser.
Audio Recorder
Record and trim audio with a live waveform.
Questions people ask
Is my video uploaded anywhere?
No. Decoding, trimming, captioning, cropping, panning, and encoding all happen in your browser tab. Your file never leaves your device. You can confirm this by opening the browser developer tools network tab while a render runs.
Why do I need WebGPU for captions?
The Whisper speech to text model is heavy. WebGPU runs it on your GPU, which is roughly 5 to 15 times faster than the CPU only WebAssembly fallback. Captions work only on browsers that expose WebGPU today, which is Chrome 113+ and Edge 113+ on desktop. Rendering itself does not need WebGPU.
How large is the Whisper model download?
The small.en model is about 240 MB compressed. It is downloaded once and cached by your browser. After the first use, caption generation works offline.
How does the timeline trim work?
The timeline is a thumbnail strip of your video. Drag the teal handle on the left to change the start, drag the handle on the right to change the end. Only the highlighted range is rendered and exported. The slider controls below the timeline do the same thing with keyboard accessibility.
How does drag to pan work?
Drag inside the preview canvas and the video moves with your finger or mouse. The crop window stays fixed on the frame, so what you are doing is choosing which part of the source footage fills that window. Quick buttons below the preview jump to Left, Right, Top, Bottom, or Center. Reset position returns the crop to dead center.
Does the video get stretched or distorted?
No. The video is scaled uniformly to cover the target frame, then the parts that overflow are cropped away. Aspect ratio is preserved. Nothing is stretched or squeezed.
What output format is produced?
MP4 with H.264 video at 1080 wide, either 1080x1920, 1080x1440, 1080x1350, or 1080x1080 depending on the selected aspect, at 30 fps. This is the safest format for Instagram Reels, YouTube Shorts, TikTok, and Facebook Reels.
Is audio preserved?
The rendered reel contains video only in this version. For clips with dialogue, generate captions so the message is preserved on screen. Audio muxing is on the roadmap.
Why is my render taking a while?
Rendering time scales with clip length. A 30 second clip on a modern laptop usually finishes in under a minute. Longer clips take proportionally longer. The first render also loads the WebCodecs muxer.
Does it work on mobile?
The UI works but WebCodecs and WebGPU support on mobile browsers is inconsistent. Use a laptop or desktop Chrome or Edge for the intended experience.
Can I edit the generated captions?
Not in this version. Captions are shown as a read-only list for reference. Copy the transcript, adjust the text in a text editor, and regenerate if the audio changed.
What platforms can I post the reel to?
Instagram Reels, YouTube Shorts, TikTok, Facebook Reels, LinkedIn, Snapchat Spotlight, and Pinterest. The output is a standard MP4 that any of them accept.
Does it cost anything?
No. ToolDeck tools are free and have no usage limit, no account, no credits, and no watermark. There is no paywall anywhere on the page.
Privacy and how we handle your video
Nothing about your video is sent to a ToolDeck server because there is no ToolDeck server involved in the render. The page loads a bundle of JavaScript and a WebAssembly module. Once loaded, your browser does the decoding, the trimming, the captioning, the cropping, the panning, and the encoding all by itself.
Whisper, the speech recognition model used for captions, runs on your GPU. The 240 MB model file is downloaded once from a public CDN and then cached by your browser. After the first use, the tool works fully offline. You can test this by disconnecting your network and generating captions again.
No account is required. There is no way for us to see what you uploaded, because nothing is uploaded. If you are curious, open the developer tools, switch to the network tab, and start a render. The only requests you will see are the initial page bundle and the one time model download.