Detecting emotions from audios using neural networks

Verification using voice biometrics

반려견 예비 보호자들을 위한 스타터 애플리케이션, 반기다: AI Server

Speaker Recognition deep learning model based on feature extraction from Mel Frequency Cepstral Coefficients

ENCM 509 - Fundamentals of Biometric Systems Design - Winter 2024

Our goal is to push the general performance of music genre recognition forward and introduce a new method for pre-processing which allows for faster experimentation and model tuning in the future. We experimented with two different musical representations: mel-spectrograms and manually extracted features.

The voice-controlled robot system features four simple commands: forward, backward, left, and right.

Voice Id Door Lock Web-App is a Speaker-Identification and Sentence-Verification using Voice MFCCs Feature and GMM

Machine Learning based Cough-Detection from Audiorecordings by using MFCCs, PCA and One-Class-SVM

EE 338 (Digital Signal Processing) - Application Assignment

This project focuses on classifying Classical Music into its sub-genres. We have performed Data cleaning, Normalization and Standardization. We have extracted Mel-Frequency Cepstral Components(MFCC), dynamic, rhythm, tonal, and spectral features from the audio files. Algorithms like K-Nearest Neighbor, Random Forest, Support Vector Machine, Mult…

Speaker verification system using MFCC’s features and MLP classifier.

A simple AI drawing interface controlled by voice commands

Reverse engineering sound.

multiple projects in speech processing

Audio classification model. PyTorch.

NLP Project for CS6120 at Northeastern University

In this project we have created a Artificial Neural Network to classify the audios along with Exploratory Data Analysis and Data Preprocessing.

A python wrapper for speech feature extractoin