Best VTuber Software

BestVTuberSoftware.md

Best VTuber software

This is a list of the most commonly used and relevant vtubing software. The "best" will always be subjective and depend on your specific requirements. Overall, the information in this list is as accurate as I could figure it out, but there might be errors or some details might become out of date. If you find anything that needs to be corrected, please let me know. You can also note it in a comment.

Additional explanations:

• iPhone means that an iPhone is basically required
• iFacialMocap support means that tracking data can be received from the iFacialMocap iPhone app
• VMC protocol means that the application can send and/or receive tracking data from other VMC protocol capable applications, allowing the combination of multiple tracking methods (e.g. VSeeFace receiving VR tracking from Virtual Motion Capture and iPhone/ARKit face tracking from Waidayo)
• Tobii means that the Tobii eye tracker is supported
• VR means that VR tracking is used and no camera based tracking is supported
• Desktop means that mouse and keyboard input is captured and can be used to animate the avatar instead of or in addition to face tracking
• Leap Motion means that hand and finger tracking through Leap Motion is supported

The order of this list is arbitrary and quite possibly biased, since I develop VSeeFace. Please also note that this list only includes information about tracking, not about other features like transparent background support, model formats and so on.

You can find another overview by Runa Yukiko here.

3D models

• VSeeFace (VMC protocol, Leap Motion)
• Tracking World (VR, VMC protocol)
• Virtual Motion Capture (VR, Tobii, VMC protocol)
• VMagicMirror (VMC protocol, iFacialMocap support, Desktop)
• Luppet (iFacialMocap support, Leap Motion)
• VUP (Leap Motion)
• SUVA (VR, Tobii)
• Hitogata
• Wakaru
• 3tene (Leap Motion)
• VDRAW (Desktop)
• Waidayo (iPhone, VMC protocol)

Live2D

• VTube Studio
• PrprLive
• FaceRig

Other

• Inochi2D (a free and open source 2D VTuber platform)
• VPupPr (a free and open source 3D VTuber software with VRM support)

Details for 3D model programs

Application	Price	Face tracking	VR	Leap Motion	Tobii eye tracker	VMC Protocol	iFacialMocap	Desktop tracking	iPhone required	Platform	Gaze tracking	Eyebrow tracking	Models	Account required
VSeeFace	free	OpenSeeFace, 66pt	no	yes	no	send, receive (from two sources)	yes	no	no	Windows	yes	yes	VRM, VSFAvatar (Unity asset bundle)	no
Tracking World	free	-	yes	no	no	send, receive	no	no	no	Windows	no	no	VRM	no
Virtual Motion Capture	free/¥300	no	yes	no	yes	send (paid only), receive	no	no	no	Windows	yes (Tobii)	no	VRM	no
VMagicMirror	free	DlibFaceLandmarkDetector, 17pt	no	no	no	no	yes	yes	no	Windows	no	no	VRM	no
Luppet	trial/¥6,000+	ULSee Lite, 30pt	no	yes	no	no	yes	no	no	Windows	no, moves with head	no	VRM	full version requires login
RiBLA Broadcast (β)	free	MediaPipe	no	no/MediaPipe	no	no	yes	no	no	Windows/Mac	?	?	VRM	no
VPupPr/openseeface-gd	free	OpenSeeFace, 66pt	no	no	no	no	yes	mouse via plugin	no	Windows/Linux	yes	yes	VRM	no
VUP	discontinued	Faceunity	no	yes	no	yes	no	no	no	Windows	yes	maps to expressions	VRM	yes
SUVA	free	-	yes	no	yes	no	no	no	no	Windows	yes (Tobii)	no	Unity asset bundle	no
Hitogata	free	Dlib, custom?	no	no	no	no	no	no	no	Windows	no?	yes	Character maker, FBX, VRM	no
Wakaru	free	Dlib, 68pt	no	no	no	no	no	no	no	Windows	yes	yes	VRM	no
3tene	free/¥2,000	Dlib, 68pt	3tenePRO only	yes	no	no	no	no	no	Windows, Mac	no	no	VRM	no
VDRAW	¥1,000	Banuba Face AR	no	no	no	no	no	no	no	Windows	?	?	VRM	?
Waidayo	free	ARKit	no	no	no	send	no	no	yes	iPhone	yes	yes	VRM	no
VKatsu	free	Dlib, 68pt	yes	no	no	no	no	no	no	Windows	no	yes	Character maker (restricted use), paid export to Virtual Cast only	no
Kalidoface 3D	free	MediaPipe	no	no/MediaPipe	no	no	no	no	no	Web	?	?	VRM	no
VTuber Maker/Live3D	free/free trial	BaiduFace	no	no	no	no	no	no	no	Windows	?	?	Character maker ($19.99)	yes
WebcamMotionCapture	trial/$1.99 per month	MediaPipe	no	no/MediaPipe	no	yes	no	no	no	Windows/Mac	no?	no?	VRM	?
Animaze	free/various fees	VisageTechnolgies, 78pt (was Hyprface)	no	yes	yes	yes	no (custom app only)	yes	no	Windows	yes	yes	various via auto import (incl. VRM)	no
Warudo	free	MediaPipe/OpenSeeFace	no	no	no	yes (multiple senders/receivers)	yes	keyboard only	no	Windows	yes	yes	VRM, WarudoMod	no

Details for 2D model programs

Application	Price	Face tracking	iPhone required
VTube Studio (Steam)	free with watermark/12.49€	OpenSeeFace 66pt, ARKit/ARCore	no
PrprLive	free/$9.99 DLC	IntraFace/GameAnimoji	no
FaceRig	$14.99+ plus $3.99 (Live2D DLC)	ULSee, 66pt	no
Animaze	free/various fees	Hyprface	no

Notes

3tene is pronounced "mitene" from Japanese みてね, not "threeteen".

Awesome list. I believe Zepeto and REALITY are honorable mentions for the mobile sphere

Needs updating. VUP supports VMC.

SUVA is now free and the developer has dropped the project.

For 2D, you may want to include Inochi2D, which is a FOSS (BSD License) 2D VTuber platform
https://inochi2d.com/

vpuppr now offers multiple tracking solutions.

• openseeface
• iFacialMocap
• meowface (android)
• vtube studio (mobile)
• mouse tracking (as an extra plug-in)

Avatar Cam on iphone

[September 2023 update]: In light of recent events might be important: Animaze has always used its own in-house indie avatar render engine for 3D and 2D and is not affected by Unity changes.

Hello Emiliana,

This is a great article.
The presented Animaze feature set though, has been inaccurate. Happy to provide the up-to-date details:

The webcam tracking solution used in Animaze was Hyprface from 2019 to 2020 (iOS) and 2020 to 2022 (Windows and iOS), but since autumn 2022 it is VisageTechnolgies (78 pt)
(Facerig was indeed ULSee from May 15 2015 to 2022).

Further Animaze details:

• Tobii - Supported
• VMC Protocol - Supported (both face and body)
• Desktop tracking (mouse and keyboard prop) - Supported
• Arkit (iOS tracking) - has been supported since the start for both 3D and Live2D, on the free tier.

Models - a lot more are supported:

• VRM (both Vroid and Perfect Sync), including 1.0 version; (I trust it is perhaps the only non-Unity-based VRM implementation out there in the vTubing space).
• GLB (Ready Player Me)
• FBX (freeform 3D)
• Hybrid 3D/Live2D avatars.
• Live2D runtimes.
• In-app 3D Character Creator (Personas).
• 80+ existing ( and largely configurable) characters
• The . avatar format (available since 2020) can be used as a container format for the above, but GLB, VRM Live2d runtimes can also be dragged and dropped in-app for a quick import, not requiring a bespoke conversion though the Animaze Editor.
• Worth calling out for creator’s sake: In the AI-art “wild west” we’re going through, the encrypted .avatar format serves as a robust option for model runtimes to be shared and usable but always protected with encryption from external interference. This prevents models/rigs created by independent creators from being reverse-engineered and fed into machine-learning algorithms for AI-based model generation without their permission and is important in preventing the abusive use that had happened 2D art and AI to extend in the rig creator space.

Extra motion:

• Sony Mocopi is supported (2023) - I trust Animaze was the 1st commercial app to fully support it
• Nvidia Maxine tracker is supported as an alternate tracker (Alpha implementation); face, but also includes some shoulder motion.
• Full Body motion capture suit is supported (Perception Neuron)
• Four different audio-based lip trackers.
• API-based custom tracking is supported (2023) - if you want to test how Fluffo world on the VSeeFaceface tracker, it is possible :).
• Has a generalized 3D mannequin retargeting engine, allowing any animations that are built for the mannequin to be retargeted on any humanoid 3D avatars (including VRM).

Other unique aspects:

• multiplayer /collab feature that can bring together all types of models: Live2D, VRM, Ready Player Me, and custom 3D characters around the same table/ in the same 3D space (launched back in Sep 2022).
• Granular individual filtering and range mapping settings on many individual face action units: 25+ configurable inputs controlling 35 individual action units.
• Spout2 live textures on 3D models (characters, props, or backdrops).
• Live2D textures on 3D models, for hybrid avatars - can also be driven by tracking inputs.

All the best,
Dragos

Thank you, I haven't really kept track of updates. I've updated the Animaze entry now.

The webcam tracking solution used in Animaze was Hyprface from 2019 to 2020 (iOS) and 2020 to 2022 (Windows and iOS), but since autumn 2022 it is VisageTechnolgies (78 pt)

Oh. The Hyprface library is still being shipped with it, so I didn't realize it changed.

Models - a lot more are supported:

Back when I added the entry, I believe it was necessary to manually convert models using the editor. Since there is now an automatic import function, I've updated it.

I was actually wondering about Animaze's Avatar format. The Animaze documentation says that it is an open standard. I've been looking around for a detailed, technical description of the format and haven't really found anything. Can the specification be found anywhere?

https://meligo.co.kr/index?lang=EN
https://store.steampowered.com/app/2138250/MELIGO_Creator/?beta=0

Thank you, I haven't really kept track of updates. I've updated the Animaze entry now.

Thank you too! Much appreciated!
Would love to discuss the avatar format with you and its mission. Can we hop on a call sometime?

The main idea with the avatar format is to help artists have a shippable /usable avatar, in a safe format, that they can author themselves, with free tools. But it needs to be a format that cannot be reverse-engineered, ripped/out, and fed into AI art training algorithms easily.
Too many open format model files are being abused by companies and fed into AI training algorithms every day, in some way stealing the artist's future from under their feet, under the disguise of "free formats and tools".

Our mission even back when we started FaceRig in 2014 was to give a customizable face for people who wanted to choose how they appear to the world, and also carve out a safe art market away from large companies' controls and give it "indie feet" for both entertainers and independent artists.
I want to congratulate you on the independent tracker you've created as that definitely further supports this mission.

You probably also know that Animaze, unlike most tools in this space, is on an indie avatar engine that we are developing (for the same reason). We have suspected that a 100% "monopoly" of big company engines like Unity on the avatar world is potentially problematic for the health of the segment, so we insisted on this crazy mission of also developing a totally indie engine for avatars, that:

a. supports as many types of avatars as we humanly can: Live2D, VRM +Vroid, Vrm+PerfectSync, GLB, and they can all sit around a table in multiplayer
b. where all external key tech aspects are secured and commercially licensed, and, once granted/bought/licensed to someone, can't be taken away at the whim of a big company board of directors ( which is the case with many free tech layers currently in the VTubing space- and this year we saw Unity make a move in that direction too).

We are trying to ensure a true safe haven, away from big companies' boards of directors reach.
I feel that your mission is similar, with the main exception that you trust that you can build that safe haven in Unity's courtyard and that they will forever play nice, and there's no risk in giving Unity that much power over it.

Here are some external insights on why that could be problematic, insights were not really visible to everyone back in 2018-2020, but are outlined in this piece by Tom Francis (Gunpoint, Heat Signature, and Tactical Breach Wizards) ) https://www.pentadact.com/2023-09-16-unitys-trap/

The world got a glimpse this year of why we considered a strategical vulnerability, that Unity's John Ricitiello would be granted a metaphorical collar around the neck of every avatar out there. What was eyebrow-raising was both the "free is not really free, was only a gain market share move, and now it is time to monetize" and "AI art is the future" angles.

To be clear I do not fault anyone for not being alarmed by this and using Unity in any way, as overall Unity is a fantastic engine, it has a TON of great features and plugins, and an indie engine can't compete with it across the board on every little thing. Using a big commercial engine is a path that definitely has to be explored, just that while exploring it is good to be aware that the myth of the "free engine" or "free tools" bestowed upon us indie mortals by some large corporations, is often a myth, it often has an asterisk that turns out to be "only free while gaining market share, and/or until a factual monopoly is established".

True avatar independence and freedom, cannot happen in a tech monopoly. Not if the avatars are all "collared" by being dependent on key tech layers from large corporations, that can be yanked away at any time, and where almost everyone has to march to the beat they dictate.

Fortunately, we started the work and were active in the market in this segment super early in 2014, way before most large companies' metaverse/avatar initiatives, and by doing so, did our part in preventing this avatar space from being patent-locked back in 2017-2019, and will do all we can to ensure there is no monopoly that could be closing this space down. Actually, we had been holding this door open for so long and in 2018-2019 we were actually wondering if something else was up, why are only big companies taking a stab at this, when all the indie initiatives will be making a real play for it? The good news is, 2020 changed all that :)

I created an open source tool that can be found here: https://github.com/ahmad88me/pytoontuber