YuanGongND/ast
A PyTorch implementation of the Audio Spectrogram Transformer (AST) for audio and speech classification tasks.
Not currently ranked — collecting fresh signals.
star history
This repository provides the code for the Interspeech 2021 paper presenting the Audio Spectrogram Transformer, a transformer-based model that processes audio spectrograms for classification. It includes pretrained models and training recipes for benchmark audio datasets including AudioSet, ESC-50, and SpeechCommands. The model can be used for downstream audio classification tasks via transfer learning.
Frequently asked
- What is YuanGongND/ast?
- A PyTorch implementation of the Audio Spectrogram Transformer (AST) for audio and speech classification tasks.
- Is ast open source?
- Yes — YuanGongND/ast is open source, released under the BSD-3-Clause license.
- What language is ast written in?
- YuanGongND/ast is primarily written in Jupyter Notebook.
- How popular is ast?
- YuanGongND/ast has 1.5k stars on GitHub.
- Where can I find ast?
- YuanGongND/ast is on GitHub at https://github.com/YuanGongND/ast.