💡 Summary
An open-source, local-first ecosystem of apps, starting with a desktop transcription tool, aiming to unify personal data in a user-owned folder.
🎯 Target Audience
🤖 AI Roast: “It's a promising ecosystem, but currently feels more like a well-branded monorepo with one app and a 'coming soon' sign.”
The app requires microphone access for transcription and may store API keys locally. The use of Tauri and local SQLite reduces cloud risks, but the build process involves a complex supply chain (Bun, Rust, Node). Mitigation: Users should audit the source and sandbox the app, especially the microphone permission.
📢 Repository Update: Whispering has become Epicenter! The original repository moved here. Everything else remains the same—same tools, same philosophy, same team.
- Whispering's evolution beyond transcription required changes to the repository's structure and branding.
- Everything else remains the same—same tools, same philosophy, same team.
- The original app lives on as Epicenter Whispering, keeping a tight focus on transcription.
- This makes room for standalone apps with complementary, but non-transcription-related features (like Epicenter Assistant).
- The new root of the Epicenter repository contains common files supporting all the apps in the ecosystem.
- Note: the old URL github.com/braden-w/whispering is now just a thin placeholder redirecting to this rebranded repository.
What is Epicenter?
Epicenter is an ecosystem of open-source, local-first apps. Our eventual goal is to store all of your data—notes, transcripts, chat histories—in a single folder of plain text and SQLite. Every tool we build shares this memory. It's open, tweakable, and yours. Grep it, open it in Obsidian, host it wherever you like. The choice is yours.
The Vision (how the ecosystem will work):
- Long-time Epicenter Whispering users may have configured numerous LLM API keys, transformations, and even local models.
- When the Epicenter Assistant is started for the very first time, all these API keys, transformations and local models seamlessly work.
- Epicenter Assistant may even access Whispering transcription history to provide personalized results out of the box.
- Of course, this works the other way, too: Whispering can take advantage of your Epicenter Assistant history to optimize transcriptions and transformations!
- In a siloed ecosystem, this requires much more effort, or may not even be possible.
Current Tools
Where We're Headed
Our vision is to build a personal workspace where you own your data, choose your models, and replace siloed apps with open, interoperable alternatives. All while preserving authenticity and being free and open source.
Quick Start
Install Epicenter Whispering
Our first app in the ecosystem. Choose your installation method:
macOS (Homebrew)
brew install --cask whispering
macOS, Windows, Linux (Direct Download)
Download the installer for your platform from GitHub Releases:
- macOS:
.dmg(Apple Silicon or Intel) - Windows:
.msior.exe - Linux:
.AppImage,.deb, or.rpm
→ Full installation guide and setup instructions
Build from Source
For developers or contributors:
# Prerequisites: # - Install Bun from https://bun.sh (run bun upgrade if there's issues) # - Install Rust and Cargo from https://www.rust-lang.org/tools/install (brew install rustup) git clone https://github.com/EpicenterHQ/epicenter.git cd epicenter bun install # Will prompt to upgrade if your Bun version is too old cd apps/whispering bun dev
Want to contribute? See CONTRIBUTING.md for fork and PR instructions.
Troubleshooting
If you encounter issues after switching branches or pulling changes (like "render_fn is not a function" errors), run from the repo root:
bun clean # Clears caches and node_modules bun install # Reinstall dependencies
For a complete reset including Rust build artifacts (~10GB, takes longer to rebuild):
bun nuke # Clears everything including Rust target bun install
Note: You rarely need bun nuke since Cargo handles incremental Rust builds well. Use bun clean first; reserve bun nuke for when things are truly broken.
Join Us
Discord Community
If you think like a generalist, build like a hacker, and value tools that respect your mind—you'll fit right in.
We're looking for contributors
If you're passionate about open source, local-first software, or are just a cracked Svelte/TypeScript developer—we'd love to build with you.
→ Read our Contributing Guide to get started
Contributors coordinate and share ideas in our Discord community.
Tech Stack
License
AGPL-3.0. Build on it. Fork it. Make it yours. Please contribute if you can.
Pros
- Strong commitment to data ownership and local-first principles.
- Clear, modular architecture for future app expansion.
- Supports multiple platforms and installation methods.
Cons
- The core 'Assistant' app is still under rebuild, limiting current utility.
- Ecosystem benefits are future-facing, not fully realized.
- Requires user to manage API keys and local models for full functionality.
Related Skills
payload
A“Payload's architecture may expose risks such as dependency vulnerabilities and potential CSRF attacks. Regular updates and using secure coding practices can mitigate these risks.”
app-monorepo
B“Powerful, but the setup might scare off the impatient.”
Disclaimer: This content is sourced from GitHub open source projects for display and rating purposes only.
Copyright belongs to the original author EpicenterHQ.
