QwenVoice v1.1.0

What's Changed

New Features

• Premium glassmorphism UI overhaul — aurora background, monochromatic liquid glass aesthetic across the entire app
• Merged Voice Design into Custom Voice tab — unified generation experience with chat-style prompt input bar
• History clip management — context menu actions, export, sort controls
• Audio player moved to sidebar — playback controls and generation status now live in the sidebar for a cleaner layout
• Native HuggingFace downloads — replaced huggingface-cli with a native URLSession download engine for faster, more reliable model downloads
• Temperature and max_tokens parameters — finer control over generation with MLX memory cleanup after each run
• Synchronous fast-launch path — faster app startup

Bug Fixes

• 10-bug fix sweep — concurrency fixes, error handling improvements, security hardening, and UX polish across Python backend, Swift services, and UI
• 5 additional verified bug fixes from codebase audit + UI polish
• Fixed generated clips not appearing in History view
• Fixed History context menu blocked by onTapGesture
• Fixed Preferences page background and toggle color
• Removed opaque list background in History view

Improvements

• English-only app (removed non-English voices for a focused experience)
• Preferences page redesign with custom ScrollView sections
• Full open source credits and README rewrite with complete codebase analysis
• Refined sidebar selection styling and project layout

Installation

1. Download QwenVoice.dmg below
2. Open the DMG and drag Qwen Voice to Applications
3. Run: xattr -cr "/Applications/Qwen Voice.app"
4. Launch Qwen Voice — first launch will set up the Python environment (~2 min)
5. Go to Models tab and download at least one model to start generating

Requires macOS 14.0+ and Apple Silicon (8 GB+ RAM).

System Requirements

• macOS 14.0 (Sonoma) or later
• Apple Silicon (M1/M2/M3/M4)
• 8 GB RAM minimum
• ~3 GB disk space for models (downloaded in-app)