A music video that looks sharp on your desktop can look soft, blocky, or over-sharpened the moment it lands on TikTok. That's not bad luck. Every platform re-compresses whatever you upload, and the export settings you choose either survive that second pass or get torn apart by it. Music video export settings are really about one thing: giving the platform's compression engine a file clean enough to compress without falling apart.
This guide covers what actually matters in an export, why each spec exists, how TikTok, Shorts, Reels, and Spotify Canvas each treat your upload differently, and how to export and upload cleanly so the version you post looks like the version you made.
Key Takeaways
- Every platform re-encodes your upload, so your original export quality only matters as a starting point, not a guarantee.
- Resolution, bitrate, frame rate, and codec are the four music video export settings that determine how much detail survives a second compression pass.
- 9:16 vertical is the format that matters most for TikTok, Shorts, Reels, and Canvas; a video shot or exported in the wrong ratio gets cropped or padded before it's compressed at all.
- A high starting bitrate gives the platform's compressor more information to work with, which is the single biggest lever for avoiding visible quality loss.
- Uploading through a browser tab, a messaging app, or a cloud link adds a second, uncontrolled compression pass before the platform even touches your file.
The export settings that actually matter
Four settings decide how a music video survives a platform re-encode: resolution, aspect ratio, bitrate, and frame rate. Codec matters too, but it's mostly locked in by whatever tool you export from. Get the first four right and the codec choice rarely causes a visible problem.
Resolution is the pixel dimensions of the file itself. Higher resolution gives a compressor more detail to preserve, but only up to the point where the platform's own encoder resizes it back down anyway. Exporting far beyond what the platform will display doesn't help; it just makes the file larger for no visual gain.
Aspect ratio is the shape of the frame. Vertical short-form platforms want 9:16. A file exported at 16:9 or 1:1 gets letterboxed, cropped, or padded with bars before it's ever compressed, which throws away picture area you can't get back later.
Bitrate is how much data is allocated per second of video. This is the setting most people ignore and the one that causes the most damage. A low starting bitrate means the platform's compressor is working from a file that's already thin on detail; there's nothing left to preserve once the second pass hits it.
Frame rate affects motion smoothness and, indirectly, compression efficiency. A mismatched frame rate (say, exporting at 24fps into a pipeline expecting 30fps) can force the platform to interpolate or drop frames, which shows up as stutter or smearing during fast cuts.
Resolution, ratio, and bitrate explained
Each of these specs exists for a platform-side reason, not an arbitrary one.
Why resolution has a ceiling. Every short-form platform re-encodes uploads to fit its own delivery pipeline, which is tuned for typical phone screens. A 1080 by 1920 vertical file is the practical target, enough detail for a platform's own compressor to work from without being wasteful. There's no benefit to sourcing a video at a resolution higher than what the delivery pipeline will actually serve.
Why 9:16 is non-negotiable for these platforms. TikTok, Instagram Reels, YouTube Shorts, and Spotify Canvas are all vertical-first surfaces. A landscape or square file gets forced into that shape one way or another, usually by cropping the sides or adding padding bars, and that transformation happens before compression, not after. If you generate with Echonos, choose 9:16 for the run (it also renders 16:9 and 1:1 for other placements). If your source is already 9:16, you skip that forced reframe entirely.
Why bitrate is the real lever. Think of bitrate as the budget the compressor has to describe motion and detail. A high-motion scene (fast cuts, particle effects, quick pans) needs more of that budget than a slow, static shot. If your export bitrate is already low, the platform's re-encode has nothing to draw from and the result looks blocky, especially in fast passages or dark scenes where compression artifacts show up first.
What happens if you ignore these. The video plays, technically. It just looks worse than what you made: softer edges, blocky shadows, visible banding in gradients, or motion that stutters during quick cuts. None of that is a generation problem. It's a spec mismatch between what you exported and what the platform's pipeline expects.
Per-platform delivery: TikTok, Shorts, Reels, Canvas
Every vertical short-form platform wants roughly the same shape, but the practical upload behavior differs enough to plan around.
| Platform | Aspect ratio | What it re-encodes hardest | Practical note |
|---|---|---|---|
| TikTok | 9:16 | Bitrate, especially on data-saver connections | Compresses aggressively on upload; start with the highest clean bitrate you can export |
| YouTube Shorts | 9:16 | Resolution scaling on lower-end playback | More forgiving of slightly lower starting bitrate than TikTok, but still benefits from a clean 9:16 source |
| Instagram Reels | 9:16 | Color and contrast, alongside bitrate | Known for punchier re-encoding; avoid over-saturating your export since Reels tends to boost it further |
| Spotify Canvas | 9:16, looping | File size ceiling for the loop itself | Canvas is a short, silent, looping clip cut from your main video, not a separate export pipeline |
The common thread: all four want a clean 9:16 file with a healthy bitrate. None of them reward exporting in the "wrong" native ratio and hoping their cropping tool does something sensible with it.
Exporting the right way
Whether your video comes straight out of an Echonos run or out of an editor like CapCut, Premiere or DaVinci Resolve, the same rules apply. Generating in the ratio you will post (Echonos lets you pick 9:16 per run) removes the reframe problem before it starts. When you export:
- Confirm the cut is finished, including any reruns of moments you were not happy with, before exporting. Running a moment again in Create or trimming it in your editor is quick, so there's no reason to export a draft you're still planning to revise.
- Export at the project's native 9:16 resolution. Don't downscale before uploading; let the platform's own pipeline handle final delivery sizing.
- Keep the file as your local master. Re-uploading the same export to multiple platforms is fine, but if you're re-exporting after a scene fix, always pull the newest version rather than reusing an older file that's missing the fix.
- Avoid intermediate re-encodes. Don't run the export through a separate compressor "just in case" unless the file is actually too large for a specific platform's upload limit; each unnecessary re-encode is a lossy pass that compounds with the platform's own compression later.
Avoiding quality loss on upload
Most of the visible quality loss artists blame on "the AI" actually happens at the upload step, not the generation step. A few specific habits cause it:
Uploading through a messaging app first. Sending your export to yourself through a chat app before uploading to the platform adds an extra, uncontrolled compression pass you don't get to configure. Upload the original file directly from your device or drive instead.
Re-saving from a cloud preview. Some cloud storage apps generate a compressed preview file when you tap to view a video; downloading that preview instead of the original asset silently swaps in a lower-quality file.
Uploading over a weak connection. Some platforms adapt encoding quality to the upload conditions they detect. A slow or unstable connection can push the platform toward a more aggressive compression profile for that specific upload.
Posting a screen recording instead of the file. Screen-recording a video playing back and uploading that recording stacks a full extra compression generation on top of the original, and it's one of the most common reasons a video looks noticeably softer than the source.
The fix for all four is the same: upload the original exported file, directly, over a stable connection, without an intermediate viewer or messaging step in between.
Codec choices and why they rarely need manual tweaking
Codec is the compression algorithm used to encode the video data itself, and it's the one export setting most artists never touch directly, because it's usually set by whatever export tool or platform you're working in rather than chosen manually. H.264 is the codec most widely supported across social platforms today, which is why most export tools default to it for anything destined for TikTok, Shorts, or Reels. A newer codec can sometimes produce a smaller file at similar quality, but compatibility matters more than marginal efficiency gains here: a file encoded in a codec a platform doesn't fully support can end up looking worse after re-encoding than a straightforward H.264 export would have, simply because the platform has to do more work to convert it first.
The practical takeaway is that codec is the setting to leave alone unless you have a specific reason to change it. Resolution, ratio, bitrate, and frame rate are the levers worth actively managing; codec mostly manages itself as long as you're exporting from a modern tool built for social video delivery.
Frame rate consistency across a project
Frame rate mismatches are a quieter cause of quality loss than bitrate or ratio problems, but they show up clearly once you notice them: motion that stutters slightly during fast cuts, or a subtle smearing effect during quick pans. This usually happens when different pieces of a project (say, a generated scene and a manually inserted clip) were created at different frame rates and combined without matching them first.
The fix is straightforward: settle on one frame rate for the whole project before you start combining scenes, and confirm every source clip matches it. If every scene comes from the same Echonos run, this is rarely an issue since scenes rendered together stay consistent by default. It becomes a factor mainly when footage from an outside source gets folded into the same timeline in your editor.
A pre-export checklist worth running every time
Before you hit export on a finished project, a short checklist catches most of the problems covered above before they become a platform-side quality complaint:
- Confirm the project is finished, including any scene-level fixes, so you're not exporting a version you'll need to redo.
- Confirm the aspect ratio is 9:16 and hasn't been accidentally altered somewhere in the editing process.
- Confirm the bitrate is set high enough for the platform's compressor to have real detail to work from, without being needlessly excessive.
- Confirm the frame rate is consistent across every scene and clip in the timeline.
- Export directly to your device or drive, and upload from that location rather than through an intermediate app or link.
Running through this in under a minute before every export is far cheaper than diagnosing a blurry or oversized upload after the fact.
Getting the export right the first time
The pattern across every platform is the same: start with a clean 9:16 file, keep the bitrate high enough that the compressor has real detail to preserve, and avoid stacking extra compression passes between your export and the upload button. None of that requires special software or a separate video editor. It requires knowing which settings matter and skipping the habits that quietly degrade a file before the platform even touches it.
If you generate the video with Echonos in 9:16, the file you download is already the right shape, so the main job left is protecting that file on the way to the upload button. For specific failure modes, the mp3 file too large to upload guide, the video too big to upload to TikTok fix, the music video too short for TikTok breakdown, and the why your music video looks blurry on TikTok walkthrough each cover one specific upload problem in depth.
FAQ
FAQ: formats, ratios, and compression
5 questions answered. Tap to expand.
Does re-encoding always lose quality, even if I do everything right?
Does re-encoding always lose quality, even if I do everything right?
Yes, technically, but the loss is only visible when the starting file gives the compressor too little to work with. A high-bitrate, correctly-ratioed 9:16 export loses far less visible detail in a platform's re-encode than a low-bitrate or wrong-ratio file does. The goal isn't zero compression; it's a clean enough source that the compression isn't visible.
Why does my video look fine on my phone but blurry after I post it?
Why does my video look fine on my phone but blurry after I post it?
Your phone is playing the original file. Once posted, you're watching the platform's re-encoded version, which went through its own compression pipeline. If the gap is large, the most common causes are a low export bitrate, an aspect ratio that forced a crop or pad, or an extra compression pass introduced during upload (a messaging app, a cloud preview, or a screen recording).
Is 9:16 required, or can I export 16:9 and let the platform crop it?
Is 9:16 required, or can I export 16:9 and let the platform crop it?
9:16 is strongly recommended for TikTok, Shorts, Reels, and Canvas. Letting the platform crop a 16:9 file means you don't control what gets cut from the frame, and the crop happens before compression, permanently discarding picture area. If you generate with Echonos, pick 9:16 for the run (16:9 and 1:1 are also available) and the problem never arises.
What's the difference between export settings and platform compression?
What's the difference between export settings and platform compression?
Export settings are the choices you make when you save your file (resolution, ratio, bitrate, frame rate). Platform compression is what happens after you upload, when the platform re-encodes your file to fit its own delivery pipeline and bandwidth targets. You control the first; you don't control the second, but a strong export gives that second pass more to work with.
Keep reading
Related Articles
MP3 File Too Large to Upload: How to Shrink It Without Wrecking the Audio
Your MP3 file is too large to upload because it's over the 40MB limit. Here's how to compress it, convert it, or re-export it without losing audio quality.
Video Too Big to Upload to TikTok: Compression That Keeps It Sharp
Your music video is too big to upload to TikTok. Learn the size and length limits to plan around, and how to compress it without losing visible quality.
Music Video Too Short for TikTok: Hitting Length Without Padding
If your music video feels too short for TikTok, learn the length sweet spots to target and how to extend or re-cut a video without adding dead space.
Written by
Hari DevanathanLead Backend Engineer
Ex-Microsoft and Senior AI/Cloud Engineer at Leidos, building NLP, OCR, vector search, and LLM pipelines that generated ~$20M annually. Owns Echonos' audio intelligence and black-box generation pipeline, including audio analysis, beat detection, and GCP infrastructure.