Project Progress

[dog-broad]

First Steps Update

• [x] Project Initialization: Set up the repository and created the initial README file.
• [ ] Data Organization: Collected and organized Telugu music into four categories: Mass, Melodies, Upbeat, and Sad.
• [ ] Feature Extraction Plan: Decided to use Librosa for audio feature extraction, focusing on MFCC, chroma features, and spectral contrast.
• [ ] Model Selection: Considering starting with a basic neural network and later exploring more complex models like CNNs.

Next steps:

• Extract audio features from the collected dataset.
• Split the data into training and testing sets.
• Start with basic model training.

[dog-broad]

Data Gathering Complete!! Songs are ready But they are in need of cleaning

Telugu_Mass: https://mega.nz/file/Hc1QiK7L#HCK7VYaRKZlvkHhLv_Y84ZgeVpbOB6TIhVWe45U_iv0

Telugu_Melodies: https://mega.nz/file/DVVD0K4J#fpA_nsHsyq8TA4OAgQFxwUcMT3rYdVstcdx1NWv47fI

Telugu_Sad: https://mega.nz/file/HRd3CS4R#WLbUmpict19P_XpxRx68_4Q-TgKi8vefBH8gffeeIDQ

Telugu_Upbeat https://mega.nz/file/fI10VKDT#6YyGNNl-YWvP1MYujWUAjPTUOmWpn96S7vMuvpvTsnw

[Noorjahan9]

MFCC (Mel-Frequency Cepstral Coefficients) is a feature widely used in music and speech processing for tasks like audio classification, speech recognition, and music genre classification. It's derived from the Mel-frequency scale, which approximates the human auditory system's response to different frequencies. We're using MFCC (Mel-Frequency Cepstral Coefficients) for our music genre classification project because: Robust Representation: MFCCs provide a robust representation of audio signals by capturing both spectral and temporal characteristics. Dimensionality Reduction: MFCCs help in reducing the dimensionality of the feature space while retaining relevant information. Invariant to Scale and Shift: MFCC make them robust to variations in pitch and tempo. This makes them suitable for classifying music across different keys and tempos. Human Auditory Perception: MFCCs are inspired by the human auditory system's response to different frequencies, making them perceptually meaningful features. MFCCs provide a compact yet informative representation of audio signals, making them valuable features for music genre classification and various other audio processing tasks.