The project’s goal is to build an automated system for real-time acoustic detection of gunshots in computer game scenarios, using deep learning. The system uses a neural network to detect gunshots.
A few stages were in the development process: First, we constructed the dataset. During this work we tested several features and chose those who proved separation between audio segments that contain gunshots to those who do not. The second stage was to find a network suitable for the needs of the project and train it using our dataset, and to perform real-time test to detect gunshots. This solution was compared to traditional classifying methods, e.g. SVM, which was applied based on the features that were used for the construction of the dataset.
The chosen solution uses the YAMnet model. This new model classifies audio signals into 521 different class types. The model was adapted for the needs of the project target using transfer learning. We succeeded to get 97% accuracy on the test set.