Voice Changer for Mozilla Hubs Successor: WebXR VR Guide
Mozilla Hubs voice changer setups worked smoothly for years — then Mozilla shut the service down in May 2024, leaving tens of thousands of virtual event hosts, educators, and open-source VR communities hunting for alternatives. The good news: the Hubs codebase is open-source, several forks are actively maintained, and a well-configured voice changer works in every browser-based WebXR successor through the same mechanism. This guide walks you through each major platform, explains the audio routing, and shows you exactly how to configure a real-time voice changer for VR presence.
TL;DR
- Mozilla Hubs closed May 2024; Hubs Cloud forks (RhizomeXR, community instances), Janus VR, and hosted options like FrameVR are the live successors.
- All WebXR platforms capture audio through the browser’s microphone API — one virtual mic configuration covers all of them.
- Set VoxBooster as Windows default recording device; browsers pick it up automatically, no extensions needed.
- Spatial audio in WebXR uses positional panning, not voice filtering — a voice changer output is treated identically to a regular mic.
- AI voice cloning brings persistent voice identity to your VR avatar across every session and platform.
- The hubs vr voice mod setup takes about three minutes once VoxBooster is installed.
What Happened to Mozilla Hubs and What Comes Next
Mozilla Hubs launched in 2018 as a WebXR experiment: browser-native VR rooms, no app download, no headset required. By 2022 it was widely used for virtual education, art galleries, developer conferences, and social spaces. Mozilla ran the hosted service at hubs.mozilla.com, while also releasing Hubs Cloud — a self-hosting option for organizations running their own infrastructure.
In May 2024, Mozilla announced the end of the hosted service. Budget constraints at Mozilla forced a consolidation of projects, and Hubs was among the cuts. Critically, the code remained fully open-source under Apache 2.0 license, which meant the community could and did fork it.
What is actively running today:
| Platform | Type | Status | Notes |
|---|---|---|---|
| RhizomeXR | Hubs Cloud fork | Active | Community-maintained; most direct Hubs successor |
| Community Hubs instances | Self-hosted Hubs Cloud | Varies | Universities, nonprofits run their own instances |
| Janus VR | Independent WebVR | Active development | Pioneer platform, different codebase from Hubs |
| FrameVR | Hosted WebXR | Active (commercial) | Browser-based rooms, no self-hosting needed |
| Spatial.io | Hosted WebXR | Active (commercial) | Higher production value, enterprise focus |
| Mozilla Hubs source | GitHub only | Archived/community | Self-hosting only, no official support |
For communities specifically looking for the open-source, self-hosted WebXR path — the closest to the original Hubs spirit — RhizomeXR and self-hosted Hubs Cloud instances are the primary options. For a broader comparison of spatial audio voice changer setups across platforms, the voice changer for spatial.io and virtual events guide covers FrameVR and hosted alternatives in depth.
How WebXR Audio Works (and Why Voice Changers Just Work)
Understanding the audio path explains why configuring a mozilla hubs voice changer — or any WebXR voice changer — is simpler than most people expect.
WebXR is a browser API that handles both visual rendering and spatial audio. When a WebXR application (like any Hubs fork) needs microphone input, it calls the browser’s getUserMedia API. The browser presents a permission dialog and, once granted, streams audio from whichever device is set as the system default recording device.
This means:
- The WebXR app does not know or care about the audio source brand. It just sees “a microphone.”
- A virtual microphone created by a voice changer is indistinguishable from a hardware microphone from the browser’s perspective.
- No browser extension is needed. No Hubs plugin. No WebRTC hack. Just set the virtual mic as the Windows default and every WebXR app picks it up automatically.
The spatial panning happens after the audio enters the WebXR world — it is applied to the transformed audio stream, so your voice effect travels through 3D space with full positional audio intact.
Setting Up a Voice Changer for Hubs Cloud Forks
The setup is the same for RhizomeXR, any community-run Hubs Cloud instance, or the official Hubs Cloud source deployed on your own server.
Step 1 — Install and Configure VoxBooster
Download and install VoxBooster on Windows 10 or 11. On first launch, point it at your real microphone as input. Choose your voice effect — pitch shift, AI voice clone, or a preset character voice.
VoxBooster creates a VoxBooster Virtual Mic device in your Windows audio stack via WASAPI. No kernel driver is installed, which means it is compatible with any browser without special permissions.
Step 2 — Set VoxBooster as Windows Default Recording Device
- Right-click the speaker icon in the system tray → Sounds
- Go to the Recording tab
- Right-click VoxBooster Virtual Mic → Set as Default Device
- Also right-click → Set as Default Communication Device
Both settings matter: browsers use the Default Device, while some communication-focused apps use the Default Communication Device.
Step 3 — Open Your Hubs Fork in the Browser
Navigate to your Hubs Cloud instance URL (e.g., a RhizomeXR room or your organization’s self-hosted instance). When prompted for microphone permission, click Allow.
Open the browser’s microphone selector if available — in Chrome, click the camera/mic icon in the address bar. Confirm VoxBooster Virtual Mic is listed as the active input.
Step 4 — Enter the Room and Test
Join a room. Use the mute/unmute button to test. Other participants will hear your transformed voice coming from your avatar’s position in 3D space.
If the browser requests a specific device: some Hubs forks expose an in-app audio device selector. In that case, select VoxBooster Virtual Mic from the dropdown directly — it overrides the Windows default for that session.
Troubleshooting: Voice Changer Not Detected in Hubs Fork
| Issue | Cause | Fix |
|---|---|---|
| Browser shows no virtual mic | Default device not updated | Restart browser after changing Windows default |
| Echo or feedback | Real mic still active | Set VoxBooster input to your mic; disable mic monitoring in Windows |
| Choppy audio | CPU load from effects | Lower voice effect quality setting in VoxBooster; close background apps |
| Participants hear nothing | Muted in-room | Check Hubs fork mute state (M key in most Hubs forks) |
| Audio OK but no spatial positioning | Joined outside VR mode | WebXR spatial audio requires entering a room in VR or spatial mode |
Using a Voice Changer in Janus VR
Janus VR is a different project from Mozilla Hubs — older, with roots in WebVR (before WebXR was standardized), and it takes a different architectural approach. Rooms in Janus are defined by JanusXR markup, a kind of spatial HTML, and the platform has an active modding and world-building community.
Audio handling in Janus VR follows the same browser getUserMedia path as Hubs forks: it captures from whatever the browser’s active microphone is. The hubs vr voice mod setup instructions above apply directly to Janus.
Janus-specific notes:
- The desktop Janus client (downloadable) has an in-app audio device selector. Point it at VoxBooster Virtual Mic directly.
- The browser-based Janus experience uses the browser’s default mic; set Windows default accordingly.
- Janus supports WebRTC for real-time voice and uses Opus codec. VoxBooster’s output is full-bandwidth audio — the Opus encoding that Janus applies happens downstream, after your voice is already transformed.
- World creators using Janus for events can combine voice effects with custom avatars for a consistent voice + visual identity.
For communities that migrated from AltspaceVR to Janus or similar platforms after the 2023 AltspaceVR shutdown, the voice changer for AltspaceVR migration communities guide covers the transition in detail, including how to preserve voice identity across platforms.
Voice Changer for RhizomeXR: The Closest Hubs Successor
RhizomeXR is the community fork of Hubs Cloud that has attracted the most active development since Mozilla’s closure. It preserves the Hubs room model, avatar system, and permission structure while adding community-driven features.
From a voice changer perspective, RhizomeXR is identical to any Hubs Cloud instance — same Hubs codebase, same WebRTC audio stack, same getUserMedia integration. Everything in the Hubs setup section above applies.
RhizomeXR-specific considerations:
- Room persistence and user accounts exist on the instance’s server — check whether the instance you are joining is community-run or self-hosted by your organization.
- Some RhizomeXR instances have custom audio processing middleware. If you experience doubled effects (your effect plus a server-side effect), check the instance’s room settings for “voice processing” or “noise suppression” toggles and disable them to avoid conflicts.
- If you are running your own Hubs Cloud fork and want to support community members who use voice changers, ensure the room’s audio settings do not apply server-side pitch or noise processing that would interact with client-side effects.
Open-Source Self-Hosted VR: Why It Matters for Voice Identity
The shift toward self-hosted WebXR communities is not just a technical migration — it represents a philosophy about ownership and persistent identity in virtual spaces. When a platform like Mozilla Hubs or AltspaceVR shuts down, user-built identities, spaces, and social connections disappear with it.
Self-hosted Hubs Cloud forks, Janus VR worlds, and similar open-source platforms give community operators control over continuity. Voice changers fit naturally into this: a persistent AI voice clone applied via VoxBooster means your voice identity travels with you regardless of which platform or instance hosts the room, because the transformation lives in your local Windows audio stack, not in any server.
Practical implications:
- Your voice persona is portable: same VoxBooster settings work in every WebXR platform, Discord call, stream, or video conference simultaneously.
- No re-registration of your “voice” with platform servers — the processing is local.
- Server operators do not handle your voice transformation, preserving some privacy in what your raw voice sounds like.
AI Voice Cloning for VR Avatar Consistency
Beyond simple pitch shift or preset effects, AI voice cloning allows you to build a stable voice character that sounds consistent session after session. This matters in community VR spaces where participants build relationships over time — a consistent voice identity reinforces avatar identity in a way that random preset effects cannot.
How it works in practice: VoxBooster’s AI voice conversion processes your real voice in real time, mapping it to the acoustic characteristics of a target voice model. The output has the speaking cadence, timing, and expressiveness of your actual speech, re-rendered in the target voice character.
For VR roleplay, virtual events, or online education scenarios, this means:
- Instructors can maintain a calm, consistent vocal persona that reduces listener fatigue across long sessions.
- VR roleplayers can sustain a character voice for hours without physical vocal strain.
- Community event hosts can develop a recognizable audio brand tied to their VR presence.
If you run regular sessions in a Hubs fork or Janus world, see our VRChat voice changer guide for advice on building a voice persona that holds up across long sessions — the techniques apply across all WebXR platforms even though VRChat uses a native client.
Spatial Audio and Voice Changer Interaction
One common worry: does a voice changer break spatial audio positioning? The short answer is no.
Spatial audio in WebXR works by applying real-time gain, panning, and distance attenuation to each participant’s audio stream based on their 3D position in the room. This processing happens at the receiving end — in the listener’s browser. The audio stream from the sender (you) is transmitted through WebRTC as a standard audio feed; what the sender is transmitting is irrelevant to how the listener’s client positions it spatially.
What this means in practice:
- Your transformed voice will appear to come from your avatar’s position in 3D space, exactly as a normal voice would.
- Voice effects that add reverb will layer on top of any WebXR reverb — use reverb effects sparingly in WebXR platforms that apply their own room reverb, or disable the VoxBooster reverb for those sessions.
- Heavy audio effects (multiple processing stages, very long echo tails) can increase CPU load. In browser-based WebXR, where the browser is also handling WebGL rendering, high CPU usage matters more than in a native app. Stick to single-pass effects (pitch shift, AI conversion, EQ) for VR sessions.
Comparing Open-Source WebXR Platforms for Voice Changer Use
| Platform | Audio Stack | In-App Device Selector | Self-Hosted | Voice Changer Compatibility |
|---|---|---|---|---|
| Hubs Cloud (any fork) | WebRTC + Janus Gateway | Some forks | Yes | Excellent |
| RhizomeXR | WebRTC + Janus Gateway | Community fork — varies | Yes | Excellent |
| Janus VR (browser) | WebRTC | No (uses browser default) | Partial | Excellent |
| Janus VR (desktop client) | WebRTC | Yes | Partial | Excellent |
| FrameVR | WebRTC | In-room settings | No (hosted) | Good |
| Mozilla Hubs source (DIY) | WebRTC + Janus Gateway | Sometimes | Yes | Excellent |
“Excellent” means the voice changer output is treated exactly as a normal microphone, no workarounds needed. All platforms listed use WebRTC for voice, and WebRTC’s getUserMedia path is the standard entry point that virtual mic devices plug into.
Browser-Specific Notes for WebXR Voice Routing
Different browsers handle microphone device selection with slight variations:
Chrome / Chromium: Most reliable for WebXR and virtual mic detection. After setting the Windows default, Chrome respects it automatically for new tabs. The address bar mic icon shows active input device — useful for confirming VoxBooster is selected.
Firefox: Supports WebXR and getUserMedia well. On first microphone prompt, Firefox lets you select the device from a dropdown — choose VoxBooster Virtual Mic directly here. You can also change the default in about:preferences (Privacy settings, Camera/Microphone section).
Edge: Shares the Chromium engine; behaves identically to Chrome for audio device selection.
Safari (macOS): Not relevant for Windows-based VoxBooster, but noted for completeness — Safari’s WebXR support is limited and virtual mic detection has been inconsistent.
For Jitsi-based video rooms embedded in websites (some education-focused Hubs implementations use Jitsi for fallback), the same virtual mic approach works. See the voice changer for Jitsi Meet guide for Jitsi-specific device routing details.
If your organization uses Matrix/Element for text and voice channels alongside a Hubs fork for 3D meeting rooms, the voice changer for Element Matrix calls guide covers the audio setup for that combination.
Setting Up for a Virtual Event in a Hubs Fork
Running a community event — a virtual conference, a social gathering, an educational session — in a Hubs Cloud fork with a voice changer involves a few extra steps beyond a casual user setup.
Pre-Event Checklist
- Install VoxBooster and configure your voice effect at least 24 hours before the event
- Run a full test in the actual Hubs fork instance — some instances have custom audio settings
- Check CPU usage during a test session; close background applications if usage is above 70%
- Disable Windows audio enhancements for VoxBooster Virtual Mic (right-click → Properties → Enhancements → Disable all)
- Turn off browser-based noise suppression if the Hubs fork offers it — it can conflict with VoxBooster’s noise suppression
- Test from the audience perspective: join as a second user on a mobile device or second computer to hear how your voice sounds to others
During the Event
- Keep the VoxBooster window open on a second monitor or in the taskbar for quick effect switching
- Use hotkeys to mute VoxBooster input if needed (faster than the Hubs mute toggle for urgent silencing)
- If the CPU usage in the browser spikes, the voice effect is the first thing to simplify — switch from AI cloning to a light pitch-shift preset
Frequently Asked Questions
Does Mozilla Hubs still work in 2026?
No. Mozilla shut down the hosted Mozilla Hubs service in May 2024. The codebase is open-source (Apache 2.0), so community-run Hubs Cloud instances and forks like RhizomeXR remain active. You need to find or host a Hubs Cloud fork to access the experience today.
Can I use a voice changer in a WebXR virtual world?
Yes. WebXR worlds like Hubs Cloud forks capture audio through your browser’s microphone permission. Set VoxBooster as your default recording device in Windows Sound settings, and the browser will pick it up automatically — no extensions or plugins needed.
What is the best Mozilla Hubs alternative in 2026?
For self-hosted open-source WebXR: RhizomeXR (maintained Hubs Cloud fork), Janus VR (WebVR pioneer with active development), and Mozilla’s own Hubs Cloud source on GitHub. For hosted options, Spatial.io and FrameVR offer similar browser-based VR meeting rooms without needing server infrastructure.
Does a voice changer work in Janus VR?
Yes. Janus VR routes audio through your OS microphone input. Point VoxBooster to your real microphone, select VoxBooster Virtual Mic as the input in Janus settings, and your transformed voice travels through Janus’s spatial audio system like any normal microphone feed.
Will a real-time voice changer add latency in WebXR spatial audio?
A well-optimized real-time voice changer like VoxBooster adds under 10 ms of processing latency. WebXR spatial audio already introduces 20-40 ms of network and decode latency, so the voice processing overhead is invisible in practice. Listeners hear your transformed voice with no noticeable delay.
Can I use a voice changer in browser-based VR on a PC without VR headset?
Yes. Most WebXR platforms including Hubs Cloud forks work in flat-screen mode directly in a desktop browser. You can participate in rooms using keyboard and mouse while still using a voice changer — no headset required.
What happened to AltspaceVR and its community?
Microsoft shut down AltspaceVR in March 2023. Much of its community migrated to VRChat, Mozilla Hubs (then still running), and open-source alternatives. With Mozilla Hubs also closing in 2024, many open-source VR communities now run self-hosted Hubs Cloud forks or Janus VR worlds. See our guide on voice changers for AltspaceVR migration communities.
Conclusion
The mozilla hubs voice changer pipeline did not disappear when Mozilla shut down the hosted service — it migrated to the open-source forks that carry forward the project’s core premise. RhizomeXR, self-hosted Hubs Cloud instances, and Janus VR all use the same WebRTC audio path that made Hubs voice changers straightforward to configure in the first place.
The technical reality is that every WebXR platform runs on the browser’s getUserMedia stack, and a virtual microphone device slots into that stack transparently. Configure VoxBooster once in Windows audio settings and the same setup covers every Hubs fork, Janus world, FrameVR room, and browser-based spatial audio experience you participate in. Your hubs vr voice mod configuration from today will be compatible with whatever open-source WebXR successor gains traction next year — the underlying mechanism does not change.
VoxBooster runs on Windows 10 and 11, creates a kernel-driver-free virtual mic, processes audio at under 10 ms latency, and includes AI voice cloning alongside standard pitch and effects presets. The 3-day free trial requires no credit card — enough time to test a full virtual event run-through in your chosen Hubs successor.
Download VoxBooster and set up your WebXR voice identity today.