In this project we used TESS voice dataset and processed it and perform emotion prediction The principal objective of this endeavor is to employ sophisticated Signal Processing methodologies for the purpose of decoding and scrutinizing emotional cues present in speech signals.
CNN and LSTM model is used in this project The system must effectively prepare the TESS dataset through various preprocessing steps, ensuring the data is cleaned, normalized, and segmented for consistent analysis. It should implement signal processing techniques such as waveplot generation, spectrogram computation, and feature extraction methods like MFCCs and Mel spectrograms to capture essential audio features.
Dataset used in this project Toronto emotional speech set (TESS) https://www.kaggle.com/datasets/ejlok1/toronto-emotional-speech-set-tess
STEPS TO RUN THIS PROJECT
first download the dataset and add your path for dataset run pip install, " -r requirements.txt", run firstly MainProgram then for GUI run Launcher
If you want documentation of this project and some help ,mail me on [email protected]