This project is a semester project for the Computer Engineering (AI, Vison and Sound) Master's program at Aalborg University, Denmark. © 2023
- Clone the repossitory
- Install requirements
- Move measurement files into designated folder structure as shown below
Audio File | Label File |
---|---|
data/measurment-2/raw/230428-006.wav | data/measurment-2/labels/230428-006.txt |
data/measurment-2/raw/230428-003.wav | data/measurment-2/labels/230428-003.txt |
data/measurment-1/amplified_jumps/230320-008-jump-1.wav | data/measurment-1/labels/230320-008-jump-1.txt |
data/measurment-1/amplified_jumps/230320-008-jump-2.wav | data/measurment-1/labels/230320-008-jump-2.txt |
data/measurment-1/amplified_jumps/230320-008-jump-3.wav | data/measurment-1/labels/230320-008-jump-3.txt |
data/measurment-1/amplified_jumps/230320-009-jump-1.wav | data/measurment-1/labels/230320-009-jump-1.txt |
data/measurment-1/amplified_jumps/230320-009-jump-2.wav | data/measurment-1/labels/230320-009-jump-2.txt |
- Run cnn_make-clips.py
- Run train_ccn.py (train_cnn.py will create numpy files that train_svm.py and train_lda.py depends on)
- If you wish to run baseline_model, then also run the make-clips.py file first.