dba78a2766
- Detection: broaden request-starters, handle contractions, allow 1-word '?' - Audio: PTT pre-roll (no clipped first word), drop Whisper silence hallucinations, greedy decoding + domain initial_prompt for faster/cleaner transcription - AI: cap spoken-answer tokens so replies return at conversational speed - Overlay: answers persist (no auto-hide); wire Ctrl+Shift+H show/hide toggle - Screen capture: freeze-frame at hotkey press (immune to focus-blur lockouts), hide selector from screen-share (NSWindowSharingNone), higher capture resolution - Stop tracking models/ and *.zip (large binaries; add to .gitignore) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
99 lines
2.7 KiB
Python
99 lines
2.7 KiB
Python
#!/usr/bin/env python3
|
|
"""
|
|
First-time setup script for Meeting Assistant
|
|
Downloads required AI models and sets up environment
|
|
"""
|
|
|
|
import os
|
|
import sys
|
|
import subprocess
|
|
import platform
|
|
|
|
|
|
def install_dependencies():
|
|
"""Install Python dependencies"""
|
|
print("📦 Installing Python dependencies...")
|
|
subprocess.check_call([sys.executable, "-m", "pip", "install", "-r", "requirements.txt"])
|
|
|
|
|
|
def download_models():
|
|
"""Download the Qwen2.5-VL vision brain (model + mmproj) into models/.
|
|
|
|
The Whisper STT model (distil-large-v3 by default) auto-downloads via
|
|
faster-whisper on first run, so it isn't fetched here.
|
|
"""
|
|
print("🤖 Downloading AI models (Qwen2.5-VL-7B + vision projector)...")
|
|
os.makedirs("models", exist_ok=True)
|
|
|
|
repo = "ggml-org/Qwen2.5-VL-7B-Instruct-GGUF"
|
|
files = [
|
|
"Qwen2.5-VL-7B-Instruct-Q4_K_M.gguf", # ~4.7 GB
|
|
"mmproj-Qwen2.5-VL-7B-Instruct-f16.gguf", # ~1.4 GB vision projector
|
|
]
|
|
|
|
try:
|
|
from huggingface_hub import hf_hub_download
|
|
except ImportError:
|
|
print("❌ huggingface_hub not installed. Run: pip install -r requirements.txt")
|
|
return
|
|
|
|
dest = os.path.abspath("models")
|
|
for fn in files:
|
|
if os.path.exists(os.path.join(dest, fn)):
|
|
print(f"✅ {fn} already exists")
|
|
continue
|
|
print(f"Downloading {fn} (this is large; behind a proxy it may be slow)...")
|
|
try:
|
|
hf_hub_download(repo_id=repo, filename=fn, local_dir=dest)
|
|
print(f"✅ Downloaded {fn}")
|
|
except Exception as e:
|
|
print(f"❌ Failed to download {fn}: {e}")
|
|
|
|
|
|
def setup_audio():
|
|
"""Test audio setup"""
|
|
print("🎤 Testing audio input...")
|
|
|
|
try:
|
|
import sounddevice as sd
|
|
devices = sd.query_devices()
|
|
input_devices = [d for d in devices if d['max_input_channels'] > 0]
|
|
|
|
if input_devices:
|
|
print(f"✅ Found {len(input_devices)} audio input devices")
|
|
for device in input_devices:
|
|
print(f" - {device['name']}")
|
|
else:
|
|
print("⚠️ No audio input devices found")
|
|
except Exception as e:
|
|
print(f"⚠️ Audio setup error: {e}")
|
|
|
|
|
|
def main():
|
|
"""Run setup"""
|
|
print("🔧 Meeting Assistant - Setup")
|
|
print("=" * 50)
|
|
|
|
# Check Python version
|
|
if sys.version_info < (3, 8):
|
|
print("❌ Python 3.8 or higher required")
|
|
sys.exit(1)
|
|
|
|
# Platform info
|
|
print(f"💻 Platform: {platform.system()}")
|
|
|
|
# Install dependencies
|
|
install_dependencies()
|
|
|
|
# Download models
|
|
download_models()
|
|
|
|
# Setup audio
|
|
setup_audio()
|
|
|
|
print("\n✅ Setup complete!")
|
|
print("Run 'python main.py' to start the assistant")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main() |