How forces23/personal_assistant works
This repository is a collection of J.A.R.V.I.S. assistant prototypes rather than one integrated application. Its clearest web workflow renders a chat interface, receives a Flask chat request, invokes the selected Llama 2 adapter, and returns text; the sampled React chatbot does not establish the client-to-server request wiring. Separate experiments cover voice capture, speech output, and standalone assistant capabilities. The legacy assistant entry points and several feature modules are represented by their tree paths, but their internal wiring was not sampled.
Main entry points: LLM_LLama2.py Llama adapter · server.py Chat endpoint · ai_chatbot_terminal.py Terminal chat · App.js React app
Components 23
Web chat
- React app
personal_assistant_web_app/client/src/App.js - Chat UI
personal_assistant_web_app/client/src/chatbot.js - Chat endpoint
personal_assistant_web_app/backend/flask-server/server.py
Voice assistant
- Assistant entry
Personal_Jarvis/jarvis-v1_0/jarvis_main.py - Assistant logic
Personal_Jarvis/jarvis-v1_0/brains.py - Keyword capture
Personal_Jarvis/jarvis-v1_0/keyword_detection.py - Assistant entry
Personal_Jarvis/jarvis-v2_0/jarvis_main.py - Assistant logic
Personal_Jarvis/jarvis-v2_0/brains.py - Speech synthesis
Personal_Jarvis/jarvis-v2_0/speaking.py
Assistant experiments
- Terminal chat
Personal_Jarvis/ai_chatbox/ai_chatbot_terminal.py - Desktop chat
Personal_Jarvis/ai_chatbox/ai_chatbot_with_ui.py - Standalone chat
Personal_Jarvis/standalone_programs/ai_chatGPT.py - Bard client
Personal_Jarvis/standalone_programs/ai_google_bard.py - Weather lookup
Personal_Jarvis/standalone_programs/weather_api.py - Web search
Personal_Jarvis/standalone_programs/google_search.py - World clock
Personal_Jarvis/standalone_programs/current_time_any_location.py - Audio translation
Personal_Jarvis/standalone_programs/translate_recording.py
AI services
- Llama adapter
personal_assistant_web_app/backend/flask-server/LLM_LLama2.py
Other components
- Web user
- Replicate API
- Google speech services
- OpenAI API
- Speaker dataset
Connections 12
- Web user → React app: opens
- React app → Chat UI: renders
- Web user → Chat endpoint: posts message
- Chat endpoint → Llama adapter: calls
- Llama adapter → Replicate API: runs model
- Llama adapter → Chat endpoint: returns text
- Keyword capture → Google speech services: recognizes audio
- Terminal chat → OpenAI API: requests response
- Terminal chat → Google speech services: synthesizes speech
- Desktop chat → OpenAI API: requests response
- Speech synthesis → Speaker dataset: loads speakers
- Standalone chat → OpenAI API: uses model