Repository navigation
Releases: Vysubs28/Jarvis-Voice-Assistant
Release list
JARVIS
🚀 JARVIS v1.0.0 – Initial Release
🆕 What's Included
Voice recognition with live speech input using speech_recognition.
Wikipedia integration for quick 2-sentence summaries on people, topics, and events.
OpenAI fallback for broader, more complex queries (uses OPENAI_API_KEY).
Google search fallback using googlesearch-python and newspaper3k to extract and summarize article content.
Text-to-speech output for natural assistant responses using pyttsx3.
Basic system commands like:
"What time is it?"
"Exit" / "Bye" to close the assistant
🔧 Setup Instructions
See README.md for full setup instructions, including installing dependencies and downloading NLTK data.
📌 Known Issues
🔐 OpenAI API quota: If you've run out of OpenAI credits, fallback to Google works, but summaries may be less precise.
🌐 Google fallback can vary: The quality of Google summaries depends on article structure.
🎙️ Microphone issues: May need to manually allow mic permissions, especially on macOS.
📦 NLTK errors: Make sure punkt is downloaded to avoid tokenizing failures.
📎 Next Up (Planned)
Add contextual memory for follow-up questions.
Improve summary parsing for Google articles.
Add custom wake word and background listening mode.