WhisperLib
by MeroSsSany
WhisperLib is a Forge library for voice recognition that runs WhisperAI on the player's own machine and posts transcribed text as Forge events.
What it does
WhisperLib is a fork of VoskLib that swaps the VOSK speech engine for WhisperAI. Recognition runs offline on the client, so microphone audio is never sent to a server. The library includes a model manager that lists available models from HuggingFace and loads them on demand, and the model is only loaded when a mod asks for it.
Other mods use WhisperLib by listening for its transcript events on the Forge Event Bus, which fire on the client side. The library has a settings screen for microphone options that VoskLib did not offer, reached from the Mods button and the Config button for WhisperLib. Players are notified when WhisperLib starts listening, and a security feature is included. The library is client-side only, so a server does not need it.
WhisperLib is published for Forge only, across six release versions 1.20.1 – 1.20.6, under the GNU Lesser General Public License v3.0 only. The source code is public. The first and latest release is 1.0.0 for 1.20.6 on the beta channel, dated 2026-03-02.
Key features
Offline voice recognition
WhisperAI transcribes speech on the player's machine, so audio is not sent to an external service.
Model manager
A built-in manager displays the models available on HuggingFace and loads them dynamically when they are needed.
Microphone settings screen
A settings screen adjusts microphone options, which VoskLib did not provide.
Event-based results
Transcripts are posted to the Forge Event Bus on the client, so other mods can react to recognized text.
Compatibility
Latest build for each game version and loader.
Only full game releases are shown. This mod publishes no snapshot builds.
Use cases
Voice-driven mods
Mod authors can build voice commands or dialogue features on top of the transcript events.
Offline speech input
Players who want speech recognition without network traffic can use the client-side engine.
Who it’s for: Mod developers who need speech-to-text on the client, and players who install a mod that depends on WhisperLib and want to adjust the microphone settings.
Questions
Does WhisperLib need to be installed on the server?
Where does the speech audio go?
Which game versions does WhisperLib support?
Where are the WhisperLib settings?
In the catalog
WhisperLib is one of 3,514 Library mods on ModsFinder, and one of 1,576 Library mods for Forge. 11,154 mods support Minecraft 1.20.6.
More from MeroSsSany
Similar mods
What the author says
An Event-based voice recognition mod that uses WhisperAI to transcript audio to text. Forked from VoskLib. WhisperLib and VoskLib are posting their results in Forge Event Bus at Client-side.
Official listing
- Modrinth page
- modrinth.com/mod/mwhisperlib
- Owner
- MeroSsSany
- Published
- 2026-03-02
- Updated on Modrinth
- 2026-03-02
- Facts verified
- 2026-10-15