• Audio Features
• Audio
• Frames
• Power Spectrum
• Filterbanks
• MFCCs
• To Scale
• Architecture
• MFCC Input
• GRU
• Model
• Training
• Accuracy
• False Positives
• 2xGRU
• Model
• Training
• Accuracy
• False Positives
• 2xMFCC Input
• GRU
• Model
• Training
• Accuracy
• False Positives
• 2xGRU
• Model
• Training
• Accuracy
• False Positives
• Mel Spectrogram Input
• GRU
• Model
• Training
• Accuracy
• False Positives
• [TimeDistributed Conv1D [10, 3], MaxPool1D]
• Model
• Training
• Accuracy
• False Positives