User repository on GitHub
A live AI soundboard powered by Jev: listens to your call (FaceTime, Discord, Zoom) and fires the perfect clip in about a second.
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Tools & CLIs
Jev Soundboard
A Python sidekick that listens to a call (FaceTime, Zoom, Discord or a stream) and fires a soundboard clip when Jev thinks the moment is funny. Jev answers two questions per 1.2-second audio slice, a Noul for react-now and a Choice for which clip, at about $0.30-0.40 per hour of chatter.
- Stars
- 4
- Language
- Python
- License
- MIT
- Reading time
- 3 min
- Published by
- Community
- Author
- Kiggsworthy
- Last verified
- Oct 2026
- Languages
- Python
- Capabilities
- Open sourceRequires API key
Highlights
- A fixed 1.2 s slicing clock instead of voice-activity detection, so overlapping talk still produces text once a second.
- One Jev call per slice with two questions in parallel: react_now (Noul) and clip (Choice).
- Code owns the rules: trigger level, 4 s minimum gap, nothing playing, and no repeats among the last 25 clips.
- 15 synthesized built-in sounds; bring your own clips via clips/index.json with a "when to use it" description.
- Costs about $0.10-0.23/hour for Jev plus roughly $0.18/hour for OpenAI transcription, shown live in the panel.
Quickstart
bash
git clone https://github.com/<you>/jev-soundboard && cd jev-soundboard
uv sync
cp .env.example .env
uv run jevboard try "he said he'd win and then fell off the map"Watch out
README reviewed 2026-10-17; the README notes the Discord bot cannot reliably hear a channel because of end-to-end voice encryption, prices are as of October 2026, and the project ships no third-party audio: clips are your responsibility.
Questions about this entry
Short answers drawn from the linked project or page.
It is an AI sidekick that listens to your call and fires the perfect soundboard clip in about a second, powered by Jev, TypeSafe's fast decision model. It works on FaceTime, Discord, Zoom or anything else with a microphone setting and ships with 15 synthesized sounds.
Jev answers two questions in parallel per audio slice: react_now, a Noul asking the probability that a sound right now makes the moment funnier, and clip, a Choice asking which sound fits best against each clip's description, or none. Code then applies the trigger rules.
Audio is committed to a transcription session every 1.2 seconds whatever happens, so text keeps flowing even without clean sentence endings; quiet slices are dropped before transcription. While Jev is answering, stale slices are skipped and only the newest moment is judged.
uv, ffmpeg (brew install ffmpeg), a TypeSafe API key for Jev and an OpenAI key for speech-to-text. The FaceTime route uses BlackHole loopback devices; Discord needs the discord extra, opus and a bot token; jevboard doctor checks keys, devices and routing.
The README estimates about $0.30-0.40 per hour of constant chatter: roughly $0.10 with the built-in sounds or $0.23 with 60 clips for Jev decisions at $0.042 per million input tokens, plus about $0.18 per hour for OpenAI's gpt-4o-mini-transcribe. Silence is skipped so real costs are lower.
No recording: audio is streamed to speech-to-text and never written to disk, and transcripts stay in memory in a rolling ~25-second window cleared when switched off. Transcript text goes to Jev and audio to OpenAI; the panel binds to localhost and is protected by a random token.
The GitHub user Kiggsworthy publishes it under the MIT license, with the built-in sounds generated by the project's own code under the same license; the repository was created 2026-10-06 and pushed the same day.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: Both hands on a piano, in real time
More Jev resources like this
A podcast chaptering tool that labels transcript windows and marks boundaries as separate local judgments, classifying content, sponsor, and housekeeping segments without generating prose for the titles.
Jev Tools & CLIs#community#python#media
Community
A local, non-autoregressive security guardrail suite powered by Convai Innovations' Laya model, returning typed decision probabilities for security inspection. It positions itself against Jev on latency and against generative LLMs on hallucination risk.
Jev Tools & CLIs#community#github#security
A CLI that checks new papers, repositories and feeds each morning against your open research questions, asking Jev Noul gates and Score dimensions through Pydantic AI's typesafe model; an LLM writes only the prose into Obsidian notes.
Jev Tools & CLIs#community#research#cli
A 2.4 GHz WiFi site-survey instrument built on a ZimaBoard 2 and an Alfa USB adapter: Python and Scapy turn captured 802.11 frames into structured RF telemetry, and Jev reads each measurement window through OpenRouter to return coverage, congestion and optimization decisions.
Jev Tools & CLIs#community#github#wifi
Community
ronibandini/jevwifianalyzer-Python
ronibandini/jevwifianalyzer-PythonA local decision engine for AI agents and software developers, built on the System One architecture in TypeSafe's Jev style and running fully offline through Ollama 0.35+. Ships a CLI, an MCP server and a Python library for error triage, diff review, shell guardrails and routing.
Jev Tools & CLIs#community#github#python
Community
beliciobcardoso/systemone_gatePython
beliciobcardoso/systemone_gatePythonModeration for communities and apps, powered by Jev: every message gets a probability for spam, scam, harassment, nsfw, doxxing and more, plus rules in plain English. Thresholds and actions are yours, and flag-only is the default.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
ohernandezdev/jevmodPython
ohernandezdev/jevmodPythonFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.