https://github.com/taylor-eos/dual-model-classifier
Abandoned learning project for audio classification
https://github.com/taylor-eos/dual-model-classifier
Last synced: about 2 months ago
JSON representation
Abandoned learning project for audio classification
- Host: GitHub
- URL: https://github.com/taylor-eos/dual-model-classifier
- Owner: Taylor-eOS
- Created: 2024-10-02T14:05:46.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-10-18T20:26:15.000Z (over 1 year ago)
- Last Synced: 2025-03-04T14:49:26.646Z (over 1 year ago)
- Language: Python
- Homepage:
- Size: 62.5 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
This was an educational machine learning project that was abandoned before it ever ran. The goal was to analyze and segment audio files into types of content. The code still contains bugs. Someone with an interest in this functionality, or an interest in learning, might be able to derive some use from it, but the dual neural network setup (one classifying 30 second segments and determining a window for a second regression model to make exact predictions) is overly convoluted and was impossible to bugfix, so I switched to [another architecture](https://github.com/Taylor-eOS/transformer-classifier).