FAQ.md

FREQUENTLY ASKED QUESTIONS

General Questions

Q: What is NYRA?

A: NYRA (Neural Your Reality Assistant) is a complete AI operating system that runs locally on your computer. It's NOT just a chatbot - it's a desktop partner with voice, avatar, memory, and deep system integration.

Q: Is NYRA completely free?

A: Yes! The software is open-source (MIT License). You only need a free Gemini API key from Google.

Q: Does NYRA work offline?

A: Partially. The UI, voice, system control, and file operations work offline. AI responses, web search, and weather require internet.

Q: What makes NYRA different from ChatGPT?

A: NYRA:

• Runs locally on your PC
• Controls your system (shutdown, open apps, etc.)
• Has memory of past conversations
• Has a visual avatar with emotions
• Speaks with voice
• Integrates with your files and screen
• Privacy-first design

---

Setup Questions

Q: Which Python version do I need?

A: Python 3.11 or higher. Python 3.13 is recommended and tested.

Q: Installation fails with "No module named X"

A: Run: pip install -r requirements.txt from the NYRA_OS directory. Make sure you're in the correct folder.

Q: How do I get a Gemini API key?

A:

1. Go to https://ai.google.dev/
2. Sign in with Google account
3. Click "Get API Key" → "Create API Key"
4. Copy the key (starts with AIza...)
5. Paste in .env file

Q: VOSK model download is huge. Is it required?

A: No, it's optional. NYRA will work fine with typing. VOSK is only needed for voice input.

Q: Do I need a GPU?

A: No. NYRA runs on CPU. It uses cloud-based Gemini AI, not local models.

---

Usage Questions

Q: How do I start NYRA?

A:

• Simple: Run python main.py from NYRA_OS folder
• Windows: Double-click start_nyra.bat

Q: NYRA doesn't respond to my commands

A: Check:

1. Is the Gemini API key set correctly?
2. Do you have internet connection?
3. Check logs/nyra.log for errors
4. Run python test_system.py to diagnose

Q: Can NYRA control my system?

A: Yes, but only when you explicitly ask. Commands like "shutdown" or "open browser" trigger system actions. Nothing happens without your command.

Q: How do I add a custom avatar?

A:

1. Get a VRM file (from VRoid Studio, Booth.pm, etc.)
2. Place it at avatar/assets/nyra.vrm
3. Update .env: AVATAR_PATH=avatar/assets/nyra.vrm
4. (Note: Full VRM rendering requires additional development)

Q: Voice doesn't work

A:

• Input (listening): Download VOSK model and set path in .env
• Output (speaking): Check audio drivers, try running as admin

Q: Can I change NYRA's personality?

A: Yes! Edit core/brain.py and modify the system_instruction to customize personality traits.

---

Technical Questions

Q: What AI model does NYRA use?

A: Google Gemini 2.0 Flash (or 1.5 Pro if 2.0 unavailable). This is configured in config/settings.py.

Q: Where is data stored?

A:

• Notes: memory/notes.json
• Logs: logs/nyra.log
• Long-term memory: memory/chroma_db/ (ChromaDB)
• All data stays on your computer

Q: Is my data private?

A: YES. NYRA:

• Stores everything locally
• Only sends YOUR MESSAGES to Gemini for AI responses
• Never tracks you in the background
• No telemetry or analytics

Q: Can I use a different AI model?

A: Currently designed for Gemini. To use others:

1. Edit core/brain.py
2. Replace Gemini client with your preferred API
3. Adjust prompt formatting

Q: Does NYRA remember conversations?

A: Yes:

• Short-term: In-memory during session
• Long-term: ChromaDB (persistent across restarts)
• Use "forget" command to clear specific memories

---

Troubleshooting

Q: Error: "Could not open requirements file"

A: You're not in the NYRA_OS directory. Run:

cd F:\Backup_Code\NYRA_AI\NYRA_OS
pip install -r requirements.txt

Q: Error: "API Key missing"

A: Edit .env and replace YOUR_GEMINI_API_KEY_HERE with your actual key.

Q: UI doesn't show or crashes

A:

• Update PyQt6: pip install --upgrade PyQt6
• Update graphics drivers
• Run as Administrator
• Check compatibility mode (Windows 8/10)

Q: TTS speaks but audio is garbled

A:

• Check default audio device in Windows settings
• Try different voice in pyttsx3 (edit voice/speak.py)
• Update audio drivers

Q: "Permission Denied" errors

A: Run terminal/command prompt as Administrator.

Q: High CPU usage

A: Normal during AI inference. Adjust system stats refresh rate in main.py (change from 2000ms to 5000ms).

---

Advanced Questions

Q: Can I add custom commands?

A: Yes! Edit core/router.py:

if 'my_command' in text_lower:
    return {'type': 'custom', 'action': 'my_action'}

Q: How to add a new service?

A:

1. Create file in services/
2. Import in core/orchestrator.py
3. Add to services dict
4. Register route in core/router.py

Q: Can I run NYRA on Linux/Mac?

A: Partially. Core logic is cross-platform, but:

• Some system controls are Windows-specific
• File paths need adjustment
• Test thoroughly

Q: How to package NYRA as .exe?

A: Use PyInstaller:

pip install pyinstaller
pyinstaller --onefile --windowed main.py

(May need adjustments for dependencies)

Q: Can multiple users use NYRA?

A: Not yet. Single-user design. Multi-user support planned for v2.0.

---

Safety & Ethics

Q: Can NYRA be used maliciously?

A: NYRA has system access, so:

• Only run trusted commands
• Don't give access to untrusted users
• Review code before executing generated scripts
• System commands require explicit user intent

Q: What data does NYRA send to Google?

A: Only your chat messages (for AI responses). No telemetry, no system info, no files unless you explicitly share them in a message.

Q: Is NYRA production-ready?

A: For personal use: YES
For enterprise: Needs security hardening, multi-user, and professional deployment

---

Support

Q: Where do I report bugs?

A: Open an issue on GitHub repository (when public)

Q: How to contribute?

A:

1. Fork the repository
2. Create a feature branch
3. Submit pull request with tests

Q: Who made NYRA?

A: Developed as a comprehensive AI OS project. Check AUTHORS file for contributors.

---

Still have questions? Check the logs at logs/nyra.log or run python test_system.py for diagnostics.