Transformer-based model for Speech Emotion Recognition(SER) - implemented by Pytorch
-
Updated
Apr 12, 2024 - Python
Transformer-based model for Speech Emotion Recognition(SER) - implemented by Pytorch
CNN Based Approach for Audio File Classification. Contains Notebooks Illustrating Data Preprocessing, Feature Extraction, Model Training, & Model Inference Workflows & Overall Pipeline
Qafar-af and Amharic voice Command Recognition project to control the movement of wheelchair
Code for the Interspeech 2021 paper "AST: Audio Spectrogram Transformer".
This project represents my research on dementia classification using audio data.
In this challenge, the goal is to learn to recognize which of several English words is pronounced in an audio recording. This is a multiclass classification task.
Classification of 11 types of audio clips using MFCCs features and LSTM. Pretrained on Speech Command Dataset with intensive data augmentation.
This repository contains the code for the paper: "DeToxy: A Large-Scale Multimodal Dataset for Toxicity Classification in Spoken Utterances"
Code for the AAAI 2022 paper "SSAST: Self-Supervised Audio Spectrogram Transformer".
Speech Classification using Continuous Attention Mechanisms
A convolutional neural network for gender classification, which achieved an F1-score of 94.3% when tested on the RAVDESS dataset. Created as postgraduate coursework, the report is included. The report also discusses Sodiq Adebiy's CNN, which I'd recommend looking at to anyone interested in emotion classification.
Wav2Vec for speech recognition, classification, and audio classification
Fall 2021 Introduction to Deep Learning - Homework 3 Part 2 (RNN-based phoneme recognition)
This repository contains code for all assignments in the Multimedia Computing and Applications (CSE563) course.
Gender Classification with different Machine Learning models, using the LibriSpeech ASR dataset.
A Python implementation of the Iterative Feature Normalization algorithm
It is a full-fetched web application.Based on sentiment classification, by using nltk library it predicts that a speech is how much toxic, sever toxic, insult, obscene, threat.
Add a description, image, and links to the speech-classification topic page so that developers can more easily learn about it.
To associate your repository with the speech-classification topic, visit your repo's landing page and select "manage topics."