Easy-to-use Speech Toolkit including Self-Supervised Learning model, SOTA/Streaming ASR with punctuation, Streaming TTS with text frontend, Speaker Verification System, End-to-End Speech Translation and Keyword Spotting. Won NAACL2022 Best Demo Award.

Audio classification with VGGish as feature extractor in TensorFlow

In this project is presented a simple method to train an MLP neural network for audio signals. The trained model can be exported on a Raspberry Pi (2 or superior suggested) to classify audio signal registered with USB microphone

Deep learning model for animal sound classification.

The official implementation of the paper "A spatio-temporal deep learning approach for underwater acoustic signals classification". In this repository, we present two new deep learning architectures based on spatio-temporal modeling for underwater signal classification.

[AAAI 2023 (Oral)] CrissCross: Self-Supervised Audio-Visual Representation Learning with Relaxed Cross-Modal Synchronicity

Learnable Gammatone Filterbank (LGTFB) and Equal-loudness Normalization (EN)

Sound Classification Dataset

Real-time version of sound_classification_demo in OpenVINO toolkit. Captures audio from microphone, do classification, and display result on the screen with illustration.

Official repository for the research article "Pruning vs XNOR-Net: A ComprehensiveStudy on Deep Learning for AudioClassification in Microcontrollers"

This repository contains an end-to-end pipeline to train a convolutional neural network (CNN) for environmental sound classification on ESC-50. Serve the model for inference using Modal + a FastAPI endpoint. - Visualize model outputs (top predictions, input mel-spectrogram, waveform, and CNN feature maps).

Classification of Sounds Using Convolutional Neural Networks

Replicating published research on sound classification using two-streamed CNN. a personal repository for the University of Bristol Applied Deep Learning unit (COMSM0018) project.

🩺♥ Heart Beat Sound Classification (HBSC) is a GraphQL API for classifying heart beats sounds in real time.

Causal depthwise separable CNN with subspectral normalization for environmental sound classification.

Kaggle competition on birdcall classification.

top 4% solution for Freesound General-purpose Audio tagging Challenge

Build a Urban sound classification using CNN.

TensorFlow implementation of Microsoft's RespireNet, aiming to classify breathing/coughing sounds of people affected by COVID19 using the Coswara dataset.

Drone malfunction finder is an end-to-end system for fault-detection-from-uav-sounds, leveraging ML-based audio analysis to identify and classify drone faults.