Deep Neural Network topologies for audio classification. This repo includes a variety of topologies, including:
Dense, Dense-LSTM,
1D-CNN, 1D-CNN-LSTM,
2D-CNN, 2D-CNN-LSTM
The implementations are powered by Python.
Specific implementations are described in the following papers:
1D & 2D CNNs vs Temporal Aggregated Feature-Based Methodologies for Audio Classification
Enhanced Temporal Feature Integration in Audio Semantics via Alpha-Stable Modeling
Experimenting with 1D CNN Architectures for General Audio Classification
You can experiment using the LVLib or BDLib Datasets:
LVLib-SMO-v1
LVLib-SMO-v2
LVLib-SMO-v3
LVLib-SMO-v4
BDLib-v1
BDLib-v2