Speech To Text Accessibility
Speech To Text Accessibility adds a client-side Whisper speech recognition feature that turns spoken words into typed text for players in Minecraft.
What it does
The mod is a bridge between people who speak and people who are hard-of-hearing, deaf or disabled, so that spoken words can be read in the game instead of typed. It relies on a JNI wrapper around whisper.cpp, a C++ port of OpenAI's Whisper automatic speech recognition model, and it can translate speech from other languages into English when a multilingual model is chosen. A speaker emoji marks text that came from this accessibility mod.
Players hold a button to speak and release it to send the transcription, which removes the need to stop moving and type during heated moments. Users choose from the available Whisper models and can load a model of their own. Settings are kept in a JSON config file, so previous choices persist between sessions. Whisper models run on the CPU without a GPU, and the smallest one needs only a modest amount of memory, while the largest is far bigger.
Speech To Text Accessibility is open source under the GNU General Public License v3.0 only. It needs Fabric API, Mod Menu and YetAnotherConfigLib (YACL) as required dependencies, and it is published for Fabric, with builds for 1.21 – 1.21.11. The latest file, 2.2.0 for 1.21.11, was published on 2025-11-08.
Key features
Push-to-speak transcription
Holding a button records speech, and the recognized words are sent as typed text when the button is released.
Selectable Whisper models
Players choose among the available Whisper models or load a custom model of their own.
Translation to English
With a multilingual model, speech in other languages can be translated into English text.
Remembered settings
Choices are saved in a JSON config file and are applied again the next time the game starts.
Gallery
Compatibility
Latest build for each game version and loader.
Only full game releases are shown. This mod publishes no snapshot builds.
Requirements
Dependencies declared by the latest file for each loader.
Required 3
How it compares
Simple Text To Speech is client-side only and open source under the GNU General Public License v3.0 or later, with files for 1.21.9 – 1.21.10, so it reads text aloud rather than converting speech. Voice Subtitles is also client-side only and open source, under the GNU Lesser General Public License v3.0 or later, for 1.21.6 – 1.21.8, and it displays subtitles for voices. Speech Heads is a Fabric mod for 1.21.1 under All rights reserved and is not open source, and its environment is client and server, which means the server must run it as well.
| Mod | Loaders | Game versions | Environment | Updated |
|---|---|---|---|---|
| Speech To Text Accessibility | Fabric | 1.21 – 1.21.11 | Client-side only | 2025-11-08 |
| Simple Text To Speech | Fabric | 1.21.9 – 1.21.10 | Client-side only | 2025-12-25 |
| Voice Subtitles | Fabric | 1.21.6 – 1.21.8 | Client-side only | 2026-08-19 |
| Speech Heads | Fabric | 1.21.1 | Client and server | 2026-07-11 |
Questions
Does Speech To Text Accessibility need to be installed on the server?
Which dependencies does Speech To Text Accessibility require?
Does the mod need a GPU to run Whisper?
In the catalog
Speech To Text Accessibility is one of 5,392 Social mods on ModsFinder, and one of 4,084 Social mods for Fabric. 19,862 mods support Minecraft 1.21.11.
Similar mods
What the author says
This mod aims to plug the communicative gaps between people who are capable of speaking and people who are incapable of hearing i.e. hard-hearing/deaf/disabled people. By relying on JNI wrapper for whisper.cpp, which is a C++ port of OpenAI's Whisper, as the intermediary base ASR model in translating what people…
Official listing
- Modrinth page
- modrinth.com/mod/stta
- Owner
- Orion_Moonclaw
- Published
- 2024-08-19
- Updated on Modrinth
- 2025-11-08
- Facts verified
- 2026-10-15