🎙️ VoskFlow Devlog — Offline Speech Intelligence Desktop App

Building VoskFlow — an entirely offline, privacy-first voice dictation engine with an Apple/VOID-inspired minimalist HUD that works across any Windows app.

Here’s everything built, debugged, and polished over the latest development sprint! 🚀

⚡ What We Built & Accomplished

1. 🌌 The “VOID” Minimalist HUD & Pill Widget

• Idle State: A discreet, 4px Warm Ochre (#DCA54C) dash docked right above the Windows taskbar.
• Active State: Hold the global push-to-talk hotkey (Ctrl + Space), and the dash smoothly expands into a floating pill with live symmetrical audio waveform visualizer bars.
• Hands-Free Auto Mode: Double-tap Ctrl to lock into hands-free streaming dictation.
• Zero Distraction: When released, it transcribes offline in milliseconds and types your text directly into whatever window/app has focus.

2. 🛠️ Frameless 6-Step Onboarding & Calibration Wizard

• Custom frameless Tauri + Webview2 window with dynamic particle starfield canvas.
• Hardware RAM auto-detection to automatically recommend the right offline model (tiny.en, base.en, small.en).
• Interactive Microphone Tuning & Voice Calibration step with live audio peak meters and immediate dictation verification.
• Seamless transition to the full system Dashboard upon completion.

3. 🧵 Bulletproof AI STT Worker & Windows Engine Architecture

• CTranslate2 & OpenMP Multi-threading Fix: Solved the infamous OpenMP CPU inference deadlock on Windows by isolating model initialization and transcription into a dedicated Python Queue worker loop.
• float32 Universal Compute Engine: Standardized compute format to ensure compatibility across all CPU architectures (Intel & AMD) without quantization crashes.
• Thread-Safe Qt Signal Dispatching: Re-architected background thread-to-GUI communication to eliminate GetDC (Invalid window handle) errors.

4. 🔕 Stealth Tray Engine & Lifecycle Management

• Persistent Background Daemon: Closing the Dashboard now cleanly minimizes VoskFlow to the system tray so speech dictation is always on standby.
• Auto-Cleanup for Stale Processes: Built automated socket & process-mutex cleanup to prevent orphan zombie backends when restarting or developing.
• Tray Bridge: Open/close Dashboard anytime directly from the Windows notification tray menu.

🧠 Tech Stack

• Frontend: Tauri v2, Rust, HTML5/CSS3 Canvas (Particle Physics), JetBrains Mono
• AI Core: Faster-Whisper (CTranslate2 offline speech models)
• Audio & System: PyAudio, NumPy, PyQt6 (Translucent GUI overlays), Pynput (Low-level global hotkeys)
• Platform: Windows 11 / 10

🔮 What’s Next

• Exportable custom voice shortcuts and automation macros.
• Local custom vocabulary & dictionary substitution rules.
• Packaging into a self-contained one-click .exe installer.

100% offline. Zero telemetry. Your voice stays on your machine. 🔒

Made initial html

🎛️ Wibei Devlog: 3D Holographic Album Art & the VOID Overhaul

Project: Wibei (3D Audio Spectrograph & Real-Time DSP Lab)
Built for: Hack Club Stardance
Creators: @EVINJSUBIN & @aamosm
Repo: github.com/EVINJSUBIN/Wibei | Live: wibei.onrender.com

🚀 Elevating the Cockpit

Audio visualization isn’t just about moving bars — it’s about tactile feedback, visual hierarchy, and immersive atmosphere. In this sprint, we overhauled Wibei into a commercial-grade glassmorphism workstation, added a 3D holographic album art core, wrote a zero-dependency ID3 tag parser, and redesigned our landing page in our signature VOID aesthetic.

💎 1. Commercial-Grade Glassmorphism Cockpit

We replaced flat panels with an ergonomic, floating dual-deck hardware layout:

• Left Deck (Ingest): Quick-switch tabs for DEMOS, STREAM (YouTube/Spotify live audio), QUEUE (playlist management), and LOCAL (drag-and-drop).
• Right Deck (Synthesis & FX): Instant geometry matrix switching (RING, WAVE, GRID, ORB), real-time biquad muffler, sub-bass drive, nightcore tempo shifters, and camera auto-orbit.
• Floating Transport Bar: Ergonomic bottom bar with real-time master VU meter bars, peak telemetry (dBFS), elapsed timers, track metadata, and volume scrubber.
• Frosted Material System: Deep obsidian glass (rgba(10, 10, 14, 0.72)) with backdrop-filter: blur(28px), subtle rim lighting, and responsive LED indicators.

💿 2. 3D Holographic Center Album Art

In our RING (Pulse) visualizer, tracks with rich artwork now come to life in 3D:

• Hologram Disc: High-def circular disc (CircleGeometry(4.2)) with dark tinted glass backing and an audio-reactive neon border ring.
• Adaptive Gyro-Core: When artwork is present, the wireframe octahedron scales outward to 1.55x, orbiting as an open crystal gimbal cage. Without artwork (or on mic input), it docks back to its native 1.0x center scale.
• Vinyl Dynamics: Spins with bass-accelerated inertia, tilts dynamically in 3D pitch/roll, and pulses to sub-bass hits.

📦 3. Zero-Dependency In-Browser ID3 Extractor

To keep local playback instant without server uploads, we wrote id3.js from scratch:

• Direct binary parsing of ID3v2.3, ID3v2.4, and ID3v1 tags via ArrayBuffer slices.
• Extracts APIC frames into instant in-memory blob: URLs for album covers.
• Parses title, artist, album, and genre instantly with zero latency.

🌌 4. The “VOID” Aesthetic & Minimalist Landing Page

• Serious Void Default: Distraction-free interstellar abyss (#050507), crisp starlight typography (#f8fafc), and titanium slate accents (#8b95a5).
• Calibrated Bloom: Re-tuned UnrealBloomPass threshold to 0.82 — eliminating blinding white glare while keeping geometry razor-sharp.
• Minimalist Landing Page: Clean terminal layout with a 1,800-star WebGL starfield and a 3D gyro-core that pulses with kinetic shockwaves on live DSP audition clicks!

💻 100% Client-Side Engine

Built entirely with vanilla JS, Three.js (r128), and native Web Audio API — zero bulky UI frameworks, locked 60+ FPS, and zero-latency 32-bit floating DSP.

🎙️ VoskFlow Devlog: Ready for V1 Ship! 🚀

Spent the last sprint squashing bugs and polishing VoskFlow (100% offline, privacy-first voice-to-text) for its V1 release!

What got done:

• 🛡️ True 100% Offline Mode: Patched faster-whisper from pinging HuggingFace API on startup. If local weights exist, it loads straight from disk—zero network required.
• 🎨 Midnight Study UI: Overhauled the web dashboard with our official palette (Espresso #1C1917 + Warm Ochre #DCA54C), serif typography, and a live status indicator.
• ⌨️ Bulletproof Push-to-Talk: Replaced basic key listeners with a low-level global hook (keyboard.hook), fixing combo hotkeys like Ctrl+Space without crashes.
• 📦 PyInstaller Monolith: Built a custom .spec bundling the models, CTranslate2 DLLs, and web UI into a clean, windowed .exe (no ugly terminal popup).

Next up: final .exe compilation and shipping! Don’t type, just speak. ✨

⚡ Wibei // 3D Audio Spectrograph & Real-Time DSP Lab

Live Site: https://wibei.onrender.com | GitHub: EVINJSUBIN/Wibei
Built for: Hack Club Stardance | Collaborators: @EVINJSUBIN & @aamosm

🌌 What We Built

Wibei is an open-source web synthesizer and 3D audio spectrograph built with Three.js WebGL and the native Web Audio API. It lets anyone route audio, manipulate live DSP filter nodes, and render acoustic harmonics in 3D space with zero latency.

✨ Key Features

• Sound Shaper DSP Suite: Real-time 650Hz Lowpass Muffler (“club bathroom” vibe), +16dB Bass Overdrive EQ, and 1.25x/0.85x Speed Modulator.
• 4 Spatial 3D Geometries: 128-bar Radial Pulse Ring, Waveform Ribbon, Isometric 3D Voxel Matrix, and the new harmonic Pulsar Sphere.
• Beat Drop Shockwaves: Sub-bass transient detector triggering expanding 3D ring particles on drops.
• Cinematic Auto-Pilot Cam: Dynamic drone camera orbiting to musical energy.
• Universal Audio Ingest: Instant lossless stems, fullscreen drag-and-drop MP3/WAV, live microphone, and YouTube stream engine.
• Hall of Fame: Live GitHub telemetry showcasing community contributors.

100% client-side WebGL & DSP running at 60 FPS.

OrbitOS 2.0 is officially ready for launch! 🚀I just completely overhauled the OS from the ground up using React, Vite, and Framer Motion. Here’s what’s new:New Taskbar: Upgraded the floating dock into a sleek, full-width frosted glass taskbar.ISS Tracker: Built a custom satellite app that pulls live telemetry and coordinates from the International Space Station every 5 seconds! 🛰️Orbit AI: Integrated a built-in smart assistant directly into the desktop.Space Theme: Fully customized the boot sequence, terminal, and styling to match a deep space aesthetic.Super hyped to finally ship this version! Let me know what you think! 🌌✨

✨ What’s new:Clean Brand Pill: Removed the gold dot from the top-left logo pill. It now features clean, minimal typography (wibei v1.0).Interactive Mini Player:Added a Minimize Button (—) on the right side of the player bar.Clicking it collapses the large deck into a floating Mini Player Pill in the bottom corner (showing the album art, track title, animated mini EQ visualizer, play/pause toggle, and expand button).Expanding it restores the full control deck seamlessly.“Getting Started” Guide Modal (Guide):Added an interactive onboarding modal accessible via the top-right Guide button.Explains how to play audio (YouTube, Demos, Mic, Uploads), how to orbit in 3D, and how to switch visualizer shapes/bloom themes.Includes a 1-click “⚡ Play Synthwave Demo” button right inside the guide for instant reviewer onboarding.Animated Mini EQ Bars: Added dynamic dancing EQ waveform bars right next to the song title when audio is playing.

Finaaly trinaed a llm and connetced it

🚀 Final Ship: Wibei is Live! 🎧✨After hours of refining the audio DSP pipeline, tweaking Three.js shaders, and perfecting the UI, Wibei is officially shipped!Try it out live here: 👉 https://wibei-production.up.railway.app/(Open it up, search your favorite track or click the 🎤 mic / 📁 upload button to watch it pulse in 3D!)

Progrees working on a new way to visulize

orbitOS devlog update!! fixing stuff for review 🛠️my ship review just came back so im changing my code to fix some stuff they mentioned!i had to go through and clean up my code files and change how my variables were named so it made more sense to read. i also took out a bunch of comments i had left in there while i was building it so the files look cleaner now.im so glad i got the feedback so i could fix it before the final finish. everything is working perfect now and im rly proud of the ISS tracker app! 🌌

Today we overhauled the core engine for commercial-grade speed and accuracy. We migrated to Faster-Whisper, which gives us flawless auto-punctuation and capitalization. To achieve a zero-latency feel, we built a background worker that streams and processes audio chunks every 500ms while you hold the hotkey, meaning the final text drops instantly on release. Finally, we engineered a dynamic disfluency filter that automatically strips out filler words like umms and ahhs before typing. Up next: connecting a local LLM to turn spoken commands into OS-level keyboard shortcuts!

Finally ready to ship

OrbitOS Devlog 2: Defying Gravity 🛸The window manager is officially online! Building a custom draggable UI from scratch was tricky but incredibly rewarding.Progress:Window UI: Designed a sleek, frosted-glass window template using HTML and CSS.The Physics: Wrote a custom JavaScript engine using mousedown, mousemove, and mouseup events. It calculates the mouse delta to perfectly reposition the window using position: absolute.Z-Index Layering: Added logic so that whenever a window is clicked, its z-index dynamically increments, ensuring the active app is always brought to the front.Next up: Building our unique mini-apps and fetching live API data! 💻

[Update] - Landing Page is Live fr 🚀

Ngl, the landing page we just dropped absolutely ate and left no crumbs. Added screenshots of our black glassmorphism pill and the bouncing voice bubbles, and the aesthetic is heavily based. Wispr Flow is literally shaking rn. No cap, the UI is serving looks. Now that the site is up, we’re ready to plug in the Vosk ML model and actually let it cook. Huge W. 🔥 https://evinjsubin.github.io/VoskFlow/landing/

OrbitOS Devlog 1: The Boot Sequence 🚀We’ve laid the foundation for OrbitOS! Aiming for a futuristic aesthetic, I implemented a deep-space, neon-accented theme.Progress:Structure: Built clean HTML with a #desktop area and a #taskbar.Design: Used layered CSS radial-gradients to generate a starfield and nebula effect from scratch (no images!).Taskbar: Designed a frosted-glass UI (backdrop-filter) with neon cyan borders and a glowing Start button.Functionality: Scripted a real-time JS clock.Next up: Tackling the window manager to make apps draggable! 🌌