qwertyzipe/swearbeeper — explained in plain English
Analysis updated 2026-05-18
Auto censor swear words while livestreaming without manual bleeping.
Clean up podcast recordings that include real-time voice audio.
Mute profanity automatically during Discord calls before it reaches other listeners.
Show a live swear counter and censorship banner as an OBS overlay.
| qwertyzipe/swearbeeper | 1038lab/agnes-ai | 3eyedtiger/video2vrcemote | |
|---|---|---|---|
| Stars | 4 | 4 | 4 |
| Language | Python | Python | Python |
| Setup difficulty | moderate | easy | easy |
| Complexity | 3/5 | 2/5 | 1/5 |
| Audience | vibe coder | vibe coder | vibe coder |
Figures from each repo's GitHub metadata at analysis time.
Requires downloading a separate Vosk Russian speech model and, for Discord routing, installing VB-CABLE.
SwearBeeper is a Windows application that listens to a microphone and automatically bleeps out swear words in Russian speech, in close to real time. It works by capturing audio into a buffer, running a copy of the sound through an offline speech recognition engine called Vosk, and checking the recognized words against a list of swear word roots and prefixes. If a match is found, a beep or a random custom sound is layered over that part of the audio before it reaches the output, using a short delay of about one to two seconds to make the timing work. Because recognition happens entirely offline through Vosk, no audio is sent over the internet. The tool is aimed at people doing livestreams, recording podcasts, or making calls on Discord who want profanity automatically muted without editing audio by hand. Users can customize the delay length, beep volume, padding around each detected word, and a whitelist of words to exclude, and they can add their own list of banned words directly in the interface. There is also support for swapping in custom .wav sounds instead of a plain beep. Beyond censoring, SwearBeeper tracks statistics: a running count of swear words caught in the current session and over all time, a ranking table of the most frequent words, and a timestamped log. It includes a system tray icon, a global hotkey to mute or unmute, a microphone level meter, and settings that save automatically. A companion OBS script connects over a local socket to show a live swear counter, a censorship banner, and a timer counting how long it has been since the last detected word. The author notes this relies on word matching rather than a precise AI classifier, so it can miss some words or trigger on others by mistake, and results depend on microphone quality and background noise. Running it requires Windows, Python 3.12, and a downloaded Vosk speech model for Russian. The project is released under the MIT license.
A Windows app that listens to your mic and auto bleeps Russian swear words in near real time using offline speech recognition.
Mainly Python. The stack also includes Python, Vosk, sounddevice.
MIT license: free to use, modify, and distribute, including commercially, as long as the copyright notice is kept.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
double-check against the repo, no cap.