tuanio / audio-classification Public

Notifications You must be signed in to change notification settings
Fork 0
Star 4

Audio Classification with AlexNet and Speech Commands dataset

4 stars 0 forks Branches Tags Activity

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 36 Commits
assets		assets
conf		conf
.gitignore		.gitignore
README.md		README.md
datamodule.py		datamodule.py
main.py		main.py
model.py		model.py
requirements.txt		requirements.txt
utils.py		utils.py

Repository files navigation

Audio Classification

Datasets

Speech Commands from torchaudio, with 35 class.

Model Architecture

Using AlexNet with Mel Spectrogram 1 channel.
Output is a softmax with 35 nodes (35 class).

How to use this respository?

Edit your configuration in conf/configs.yaml
Train model using python main.py -cp conf -cn configs

Additional

You guys should install PySoundFile on windows or sox on linux, for torchaudio I/O backend.

About

Audio Classification with AlexNet and Speech Commands dataset

pytorch speech-recognition alexnet audio-classification speech-commands pytorch-lightning

Report repository

Releases

No releases published

Packages

No packages published

Languages

Python 100.0%