Recognized non-speech human sounds such as:
Clapping 👏 Footsteps 🦶 Brushing Teeth 🪥 Drinking Sipping 🧃 Laughing 😂 Breathing 🌬️ Crying Baby 😭 Coughing 🤧 Snoring 😴 Sneezing 🤧
Test the model at: link (hugging face space)
ESC-50 dataset. Using only non-speech humman sounds.
- Random Forest
- SVM
- XGBoost
- CNN on Mel spectrogram
- CNN on Mel spectrogram + Deltas
- CNN on Mel spectrogram + Deltas + Augmentations
- CNN on Mel spectrogram + Deltas + Trimmed Audios
- CNN-LSTM