 赵昌
|
8dddc3fd48
|
refactor(acoustic): Add docs, gitignore, cleanup deps, enhance WAV reader
- Add README.md with module overview, build instructions, ROS integration
- Add .gitignore for build artifacts, IDE files, Python cache
- Remove kiss_fft references from build_demo.bat and CMakeLists.txt
- Enhance wav_file_source: support 8/16/24/32-bit PCM and 32-bit float WAV
- Verify 20/20 validation accuracy maintained after all changes
|
1 day ago |
|
赵昌
|
e589cdd874
|
feat(acoustic): Recover full acoustic analyzer module with buildable demo
- Recover all 50+ source/header/script files from conversation history
- Regenerate synthetic binary dataset (ambient vs threat)
- Retrain CNN-GRU classifier: 15 epochs, 90.4% train / 100% val accuracy
- Export ONNX model with MaxPool1d workaround for opset 13 compatibility
- Fix recovered file API mismatches (AudioBuffer, Pipeline, demo_offline)
- Replace kiss_fft with inline DFT in fft_utils.cpp (no external lib needed)
- Add missing #include <memory> to PIMPL headers
- Fix ONNX input layout: [batch, time_frames, n_mels]
- Downgrade ORT_API_VERSION 20 -> 17 for local DLL compatibility
- Simplify build_demo.bat for Windows MinGW (no yaml-cpp dependency)
- Verify demo_offline.exe on validation set: 20/20 correct classifications
|
1 day ago |
p82feo7wg