I am planning to add other restreaming options such as RTMP or HLS. So far, my test stream is running for more than 10h and did not have a single dropout.
I built TinyIce as a vibing side project to spin up an Icecast2-compatible server in seconds, because I was frustrated with IceCast. One static Go binary, embedded assets, auto-generated creds on first run, built-in ACME (Let’s Encrypt), relays, multi-tenant admins, Prometheus metrics, and a modern web UI.
It's a very nice project. Me and some friends toyed with the idea of running our own IceCast server, as a way to introducing new music to each other. We eventually gave up, exactly due to frustrations with setting up and running IceCast.
I think it's really neat how you managed to include ACME, a nice UI and even the Prometheus metrics.
Get onto Music League for introducing new music to each other.
Someone setup a league at work, and it's been one of the best (albeit unintentional) team bonding exercises I've ever come across (I've not come across many). So much so that three people who have left the company still participate in the league.
It unfortunately it's linked to and requires the use of Spotify, for those who are ideologically opposed (which also means I can't submit King Gizzard and the Lizard Wizard songs anymore).
I built this because I wanted a meeting tool that didn't compromise privacy.
It’s a application that handles transcription locally and turns conversations into structured knowledge.
It has a live mode where you can also ask it questions during the meetings and a grounded research mode.
The performance of whisper.cpp allows me to transcribe 1h of meeting in just 20-30s on a Mac Pro M4.
Why it’s different:
* Local-First: Transcription via whisper.cpp happens 100% offline.
* Actionable: It doesn't just summarize; it automatically creates GitHub/GitLab issues from
meeting action items.
* Integrated: Generates structured Obsidian notes (with Mermaid.js graphs) and sleek,
HTML reports.
* Live Copilot: Press [Space] to ask the AI questions about the current meeting context in
real-time.
The Stack:
C++17, whisper.cpp, FTXUI (for the TUI), PortAudio, and libcurl (supporting Gemini, OpenAI, or
local Ollama).
I have C++ and Go projects which have 10k+ lines of code and which have compiled and worked on first try. Are you using a custom CLAUDE.md to instruct it? You can use a global one and also a per-project CLAUDE.md to give it project specific instructions.
Just a day ago, it wrote a ZeroTier userspace network backend for the QEMU/KVM virtualisation platform, which allows to use ZeroTier networks as virtual ethernet devices with proper L2 (e.g. internet wide switch) with VMs - no matter if it is Win95, Linux, QNX, ..
* Multiple AI Providers: Choose between Google Gemini, OpenAI (cloud) or Ollama/DeepSeek (local) for translations
* Multi-File Support: Process multiple files with automatic deduplication to save API calls
* Incremental Caching: Only translates new or modified strings, dramatically reducing API calls
* Batch Processing: Intelligently batches translations for optimal performance
* Path Preservation: Maintains exact JSON structure including nested objects and arrays
* Cross-Platform: Works on Windows, macOS, and Linux with automatic cache directory detection
* Developer Friendly: Built-in performance statistics and progress indicators
* Cost Effective: Minimizes API usage through smart caching and deduplication
* Language Detection: Automatically detect source language instead of assuming English
* Multiple Target Languages: Translate to multiple languages in a single command
* Translation Metadata: Optionally include translation details in output files for tracking
* Dry Run Mode: Preview what would be translated without making API calls
* Format Preservation: Maintains URLs, emails, dates, numbers, and template variables unchanged
reply