| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| Radio_transcription_tool.v3.7_Release.zip | 2025-09-05 | 156.2 MB | |
| Radio Transcription Tool Version 3.7-First modular version source code.tar.gz | 2025-09-05 | 215.1 kB | |
| Radio Transcription Tool Version 3.7-First modular version source code.zip | 2025-09-05 | 228.7 kB | |
| README.md | 2025-09-05 | 2.4 kB | |
| Totals: 4 Items | 156.7 MB | 0 | |
v3.7.0 (2025-08-31)
MAJOR REFACTORING: Modular Structure Implementation
ARCHITECTURE OVERHAUL: - Complete refactoring from monolithic to modular structure - Split 3,700+ line single file into 9 focused modules - Implemented proper separation of concerns following programming standards - Enhanced maintainability, readability, and scalability
NEW MODULAR STRUCTURE: - main.py: Main entry point and application initialization - config.py: Centralized configuration and constants management - logging_config.py: Unified logging setup and helper functions - utils.py: Common utility functions and file operations - gui.py: Complete GUI interface with menu system and user controls - audio_processing.py: Audio recording, processing, and quality enhancement - transcription.py: OpenAI Whisper integration and keypoint extraction - phrase_filtering.py: Advanced phrase processing and filtering logic - app.py: High-level workflow coordination and result management
BENEFITS: - Much easier to locate and modify specific functionality - Individual modules can be developed and tested independently - Better support for team development and code collaboration - Easier to extend with new features and capabilities - Improved code organization and documentation
TECHNICAL IMPROVEMENTS: - Enhanced error handling and debugging capabilities - Better memory management and resource utilization - Improved import structure and dependency management - Comprehensive documentation for each module - Maintained full backward compatibility with existing functionality
BUG FIXES: - Fixed critical bug where phrases were being extracted but not displayed in results - Resolved deduplication logic that was incorrectly processing single words as phrases - Phrases are now properly separated from words during deduplication process - Added comprehensive debugging to track phrase extraction and display pipeline - Fixed issue where "21 keypoints (20 words, 1 phrases)" would show "No significant phrases encountered"
USER EXPERIENCE: - Phrases are now properly displayed in both GUI popup and output files - More accurate reporting of extracted content (words vs phrases) - Better visibility into what content was actually found during transcription - Same familiar interface with improved underlying architecture