How kal51411/alicia works
Alicia is a personal voice assistant with CLI and desktop entry modes. The principal workflow accepts typed or spoken input, transcribes speech when needed, dispatches commands to AI, search, memory, or system capabilities, then presents a response with optional speech. Its major internal layers include voice processing, command/AI logic, system tools, configuration, and local persistence. The README documents VAD and SQLite, while the tree and sampled storage code show JSON-backed persistence; VAD wiring and some integration details are not established by the excerpts.
Main entry points: action.py Command Dispatcher · cli.py Terminal Assistant · mainwindow.py Desktop Window · thread.py GUI Workers
Components 24
User Interfaces
- Application Entry
src/main.py - Terminal Assistant
src/cli.py - Desktop Window
src/gui/mainwindow.py - GUI Workers
src/gui/thread.py
Voice Processing
- Voice Activity Detection
src/core/voice/vad.py - Speech Recognition
src/core/voice/stt.py - Wake Word Matching
src/core/voice/fuzz.py - Speech Synthesis
src/core/voice/tts.py
Command Intelligence
- Command Dispatcher
src/core/cmd/action.py - Gemini Conversation
src/core/logic/model.py - Text Summarization
src/core/logic/summary.py - Vision Logic
src/core/logic/vision.py - Web Search
src/core/cmd/search.py
Platform and State
- System Commands
src/core/cmd/system.py - Cross-platform Actions
src/core/utils/crossplatform.py - Path Utilities
src/core/utils/crosspath.py - Application Settings
config/settings.py - JSON Persistence
database/manager.py
Other components
- User
- Google Gemini
- Hugging Face
- Web Services
- Operating System
- Local JSON Files
Connections 39
- User → Application Entry: starts
README.md - Application Entry → Terminal Assistant: launches default
src/main.py - Application Entry → Desktop Window: launches GUI
src/main.py - Desktop Window → Terminal Assistant: falls back
src/main.py - User → Terminal Assistant: types or speaks
src/cli.py - User → Desktop Window: interacts
src/gui/mainwindow.py - Terminal Assistant → Speech Recognition: transcribes input
src/cli.py - GUI Workers → Speech Recognition: captures speech
src/gui/thread.py - GUI Workers → Wake Word Matching: checks phrase
src/gui/thread.py - Terminal Assistant → Wake Word Matching: checks wake word
src/cli.py - Desktop Window → GUI Workers: starts workers
src/gui/mainwindow.py - Terminal Assistant → Command Dispatcher: dispatches command
src/cli.py - GUI Workers → Command Dispatcher: executes command
src/gui/thread.py - Command Dispatcher → Gemini Conversation: requests response
src/core/cmd/action.py - Gemini Conversation → Google Gemini: sends prompt
src/core/logic/model.py - Command Dispatcher → Text Summarization: summarizes text
src/core/cmd/action.py - Text Summarization → Gemini Conversation: uses fallback
src/core/logic/summary.py - Text Summarization → Hugging Face: requests summary
- Command Dispatcher → Vision Logic: delegates vision
src/core/cmd/action.py - Command Dispatcher → Web Search: delegates search
src/core/cmd/action.py - Web Search → Web Services: queries web
- Command Dispatcher → System Commands: runs system action
src/core/cmd/action.py - GUI Workers → System Commands: samples metrics
src/gui/thread.py - System Commands → Operating System: reads or controls
src/core/cmd/system.py - Command Dispatcher → Cross-platform Actions: opens target
src/core/cmd/action.py - System Commands → Cross-platform Actions: uses platform helper
src/core/cmd/system.py - Desktop Window → Path Utilities: uses path helper
src/gui/mainwindow.py - Command Dispatcher → JSON Persistence: stores and reads notes
src/core/cmd/action.py - Terminal Assistant → JSON Persistence: reads user data
src/cli.py - Desktop Window → JSON Persistence: reads user data
src/gui/mainwindow.py - JSON Persistence → Local JSON Files: reads and writes
database/manager.py - Terminal Assistant → Speech Synthesis: speaks response
src/cli.py - GUI Workers → Speech Synthesis: uses speech output
src/gui/thread.py - Desktop Window → Speech Synthesis: uses speech output
src/gui/mainwindow.py - Speech Recognition → Application Settings: reads voice options
src/core/voice/stt.py - Speech Synthesis → Application Settings: reads voice options
src/core/voice/tts.py - Gemini Conversation → Application Settings: reads AI options
src/core/logic/model.py - JSON Persistence → Application Settings: uses database paths
database/manager.py - Voice Activity Detection → Speech Recognition: speech gating