Thanks to visit codestin.com
Credit goes to github.com

Skip to content
View ssh-den's full-sized avatar
44 followers · 31 following

Block or report ssh-den

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ssh-den/README.md

Hey, I'm Den 👋

I build local-first, privacy-friendly software — speech and audio processing, LLM tooling, Home Assistant integrations, and automations that take repetitive work off your hands.

Stack: Python · JavaScript/HTML/CSS · REST APIs · Docker · whisper.cpp / MLX / LLM APIs

Featured projects

PadRelay

Python networking low latency protocol design

Stream gamepad input across machines. Captures input on any OS and emulates a virtual Xbox 360 / DualShock 4 controller on a remote Windows PC — TCP or UDP transport, challenge/response authentication, heartbeats, and rate limiting. Published on PyPI.

Moonlight Voice

Python JavaScript Home Assistant REST API

A local audio response service for Home Assistant, shipped as an add-on with a hand-built web UI (vanilla JS/HTML/CSS, served through authenticated Ingress). Serves preloaded audio clips via an OpenAI-compatible TTS endpoint or Home Assistant's native TTS platform — fast, predictable voice automations.

Simple Meeting Transcriber

Python whisper.cpp audio processing

Turns meeting recordings into readable text, 100% offline, using whisper.cpp. Speaker separation for stereo recordings, silence-based chunking, parallel processing.

council-debates

Python LLM orchestration OpenAI-compatible APIs

A modular Python package for role-based LLM debates: independent roles answer the same prompt, an optional ranker scores them, and a chairman model synthesizes the final answer. Works with OpenAI, OpenRouter, Ollama, LM Studio, vLLM, and any OpenAI-compatible endpoint.

More tools

Project Description
fish-s2-mlx Local text-to-speech and voice cloning on Apple Silicon via MLX, with a Gradio UI
codex-quota Isolated Codex CLI profiles and quota monitoring, with JSON output and a live terminal dashboard
mp3-rss-downloader Batch-download MP3s from RSS feeds
zed-keep-a-changelog-snippets Zed Editor snippets for Keep a Changelog–style changelogs
ComfyUI-RUAccent ComfyUI custom node for automatic Russian stress accenting with custom dictionaries
yt-timecode-downloader YouTube downloader that extracts and ranks timestamped comments
simple-twitch-downloader Small command-line wrapper around yt-dlp for downloading Twitch VODs.

Get in touch

Questions, ideas, or opportunities — my inbox is open.

→ [email protected]

Pinned Loading

  1. PadRelay PadRelay Public

    Stream your gamepad across machines. PadRelay captures local input from any OS and emulates a virtual gamepad on a remote Windows PC.

    Python 2

  2. moonlight-voice moonlight-voice Public

    Moonlight Voice is a local audio response service for Home Assistant. Instead of generating text-to-speech, it returns a preloaded MP3 or WAV file.

    Python 1

  3. zed-keep-a-changelog-snippets zed-keep-a-changelog-snippets Public

    Markdown snippets for writing changelogs in Zed Editor that follow Keep a Changelog structure

    1

  4. mp3-rss-downloader mp3-rss-downloader Public

    Downloading mp3 from rss feeds

    Python 6 2

  5. simple-meeting-transcriber simple-meeting-transcriber Public

    Turn meeting recordings into text - 100% local, no API needed, offline transcription using whisper.cpp

    Python 1

  6. fish-s2-mlx fish-s2-mlx Public

    Local Fish Speech S2/S2 Pro text-to-speech and voice cloning app for Apple Silicon via MLX (unofficial).

    Python 1