Is Aegisub Clipper still the best way to clip Vtuber streams?

It is still the best way to produce hardsubbed clips, because Aegisub is a real subtitle editor and Clipper cuts frame-accurate multi-segment clips from lines you choose yourself. What it does not do is find the moment inside a multi-hour archive or reframe the result to vertical 9:16, which are the two steps that consume the most time when you clip regularly.

How do I clip a Vtuber stream without installing Aegisub?

Use a tool that takes the stream URL and returns finished clips. The trade-off is subtitles: automatic clippers generally do not produce the hardsubbed, typeset translations that Aegisub is built for. If your clips are mainly reaction or gameplay moments that need no translation, skipping Aegisub costs you nothing.

Can I use both Clipper and an automatic clip tool?

Yes, and they combine cleanly because they solve different steps. Let an automatic tool find the moments and produce the vertical cut, then bring that file into Aegisub if you want subtitles burned in. Nothing about the automatic step prevents subtitling afterwards.

Why do automatic tools crop my Live2D model out of the frame?

Most automatic vertical reframing detects a human face and centres the 9:16 crop on it. A Vtuber's avatar is rendered art, is often a small corner box over full-screen gameplay, and competes with game characters the tracker may lock onto instead. The fix is placing the layout deliberately rather than tracking a face.

Does clipping a stream automatically break the agency guidelines?

No. Guidelines govern what you publish, not which software produced it. The conditions are the same either way: credit and link the source stream, never clip members-only content, and check whether the agency requires registration before you monetize.

Aegisub Clipper vs Automatic Vtuber Clipping

The top Google result for "vtuber clipping tool" is not a product — it is documentation for an Aegisub script that cuts hardsubbed clips by hand. That is a legitimate answer. This page is for the clipper who read it and wondered which parts of that workflow are worth keeping and which are worth automating.

Search "vtuber clipping tool" and the top result is not a product page — it is documentation for Clipper, an Aegisub automation script that cuts hardsubbed clips out of a longer video. That is a real answer, and for a lot of clippers it is the right one. It is also the reason this page exists: if you landed on that guide and thought "this is a lot of steps", here is what the manual route actually costs, what it is genuinely better at, and where automating part of it makes sense.

Disclosure up front: we make VTubeClip, an automatic Vtuber clip maker. We are not going to pretend Clipper is bad. It does one thing extremely well that we deliberately do not do at all.

What the Aegisub + Clipper route actually is

The workflow, as documented by its author:

Install Aegisub, drop clipper.lua into %APPDATA%\Aegisub\automation\autoload, and place ffmpeg.exe and ffprobe.exe in an autoload/bin folder.
Download the VOD yourself (the guide suggests youtube-dl; most people now use yt-dlp).
Open the video in Aegisub and save a subtitle file — Clipper only reads saved changes.
Subtitle the scenes you want, translating as you go.
Mark each segment as a commented line: its start and end times define the cut.
Select those lines and run Clipper. It burns the subtitles in and joins the segments into one video.

The output is a hardsubbed clip built from frame-accurate segments you chose yourself. It costs nothing, runs offline, and nothing about your footage leaves your machine.

What Clipper is better at — genuinely

Hardsubbed translation. If your channel's value is translating a Japanese stream for an English audience, subtitles are the product. Aegisub is a proper subtitle editor: styling, timing, typesetting, karaoke. Nothing automatic comes close.
You decide the exact frames. A commented line is a precise in/out point. No detector second-guesses you.
Free and offline. No account, no credits, no upload.
Multi-segment edits. Stitching several moments from across a stream into one clip is a first-class feature, not a workaround.

If you make horizontal, subtitled clips for YouTube and you already know which moments you want, the manual route is not a compromise. It is the correct tool and you can stop reading here.

The two parts it does not do

Both are the parts that scale badly.

1. Finding the moment

Clipper starts from "you already know the timestamp". Getting to that timestamp means watching a 3-to-8-hour archive, or scrubbing it hoping to land on something. Measured across a normal clipping session, this is the single biggest time cost — larger than subtitling, far larger than the cut itself. It is also the step that makes people quit after a handful of clips.

There are ways to shorten it that do not involve watching everything: the chat replay's message rate, the viewer graph, or audio analysis — laughter, screams, sudden loudness jumps, changes in vocal pitch are all measurable without a human in the loop.

2. Vertical 9:16 for Shorts and TikTok

Clipper outputs the source aspect ratio. Short-form platforms want 9:16, which means throwing away most of the width, and for a Vtuber that is not a crop you can safely automate with a face tracker: the subject is a Live2D or 3D model, usually parked in a small corner box over full-screen gameplay. Face detection either misses it or locks onto a game character, and the model ends up half out of frame. The two layouts that actually work — a game/model split, or the model filling the frame — have to be placed deliberately. We wrote that up separately in game vs model layouts.

Using both, which is what we would actually suggest

These tools do not overlap much, so the sensible split is obvious once you see it:

Automate finding + vertical framing. Paste the stream URL, get back vertical clips of the moments worth looking at. That kills the hours-long search and the crop problem in one step.
Keep subtitles human. Translation and typesetting in Aegisub is where a clip channel's actual value lives — and, incidentally, it is also what makes YouTube treat a clip as transformative rather than reused content.

VTubeClip covers the first half: it finds candidate moments by audio analysis and renders vertical clips that keep the model and the gameplay both on screen. It adds no subtitles at all and uses no generative AI — it cuts and frames the streamer's own footage, nothing is synthesised. Take the output into Aegisub if you want subs on it.

If you want the automatic side compared against the other options, that is the tool comparison. If you want the manual side done well, the Clipper guide is the right place and we are not going to reproduce it here.

Before you clip anything, either way

Tool choice does not change the rules. Credit the source stream with a link, never clip members-only content, and check the agency's derivative-work guideline before monetizing — several require the clipper to register first. We cover that in can you monetize Vtuber clips?, with agency-specific steps for Hololive and Nijisanji.