The document summarizes research on extracting essential features from the aaroh and avroh of the Indian classical raga Yaman. It discusses extracting various time domain, spectral, cepstral and other features using audio signal processing and machine learning techniques. Time domain features like zero crossing rate and autocorrelation are extracted. Spectral features including the spectrogram, mel-spectrogram, spectral centroid, roll-off and others are analyzed. Cepstral features such as MFCC are also discussed. The goal is to extract relevant features that can be used for audio classification, recognition and other tasks related to Indian classical music.