- python >= 3.9, torch >= 2.0
- need install clip, transformers, audiocraft
- data: about the dataset and vocab, in this project, we use LAION as the dataset, the link is: https://laion.ai/blog/laion-400-open-dataset/
- dataProcess
- getVocab.py: create vocab
- precess.py: precess data
- getImgTextData.py: create dataset about image and text features
- getTextFeature.py: create textfeature dataset
- models: about models to upload in my github
- module: about how to design models
- image2Text.py: create Image2Textfeature model
- textFeature2Text.py: create textFeature2Text model
- loadMusicGen.py: load musicGen model
- image2music.py: main function that input a image then generate music
we will update this project in github if we have some updates, the link is:https://github.com/Rachel1111