Voice Note Thumbnail Generator

A Streamlit application that converts voice notes into visual thumbnails using Groq's high-speed AI inference with Whisper 3 and Llama 70B.

Features

• Upload audio files (MP3, WAV, M4A, FLAC)
• Ultra-fast transcription using Groq Whisper 3
• Smart quote extraction using Llama 3.1 70B Versatile
• Manual text size control with real-time adjustment
• High-resolution thumbnails (1024x1024) with:
  • Beautiful gradient backgrounds
  • Real emoji images from Twemoji
  • Actual quotes from your voice notes
  • Size-based importance hierarchy

Setup

1. Install dependencies:

pip install -r requirements.txt

2. Get your Groq API key:

   • Visit https://console.groq.com/keys
   • Create a free account and generate an API key

3. Set up environment:

export GROQ_API_KEY="your-groq-api-key-here"

4. Run the application:

streamlit run app.py

Usage

1. Upload an audio file
2. Click "Generate Thumbnail"
3. Use the Text Size Multiplier slider to adjust text size
4. Download your custom thumbnail

Technology Stack

• Transcription: Groq Whisper Large v3 (ultra-fast, high-accuracy)
• Quote Analysis: Llama 3.1 70B Versatile (powerful language understanding)
• Emoji Rendering: Twemoji (high-quality emoji images)
• Text Rendering: Cross-platform font handling with wrapping
• Interface: Streamlit with real-time controls

Performance

• Lightning Fast: Groq's LPU inference delivers 276+ tokens/second
• High Quality: 1024x1024 resolution thumbnails
• Smart Caching: Emoji images cached for faster regeneration
• Real-time Updates: Instant text size adjustments