AlphaTex exporter #1123
Replies: 4 comments
-
AlphaTexExporter was implemented in an older version but has been completely removed. https://www.alphatab.net/docs/guides/breaking-changes-097-098/ |
Beta Was this translation helpful? Give feedback.
-
Some time ago I prioritized that a Guitar Pro exporter would be more valuable than an alphaTex importer. I think the alphaTex adoption might be rather low, and it also does not support at this point all elements so that a full export of all information would be possible. I added the wish to the roadmap: https://github.com/orgs/CoderLine/projects/12/views/1 Depending on your plans I could check if I could provide a preliminary exporter for your needs. I'm curious to hear what your plans are. |
Beta Was this translation helpful? Give feedback.
-
Thank you for your reply.
Specifically how we would like to use alphaTex is as follows.
By doing these things, we can turn scores that no one sees anymore due to image or pdf format into easy-to-consume content again. While alphaTex is very easy to use, it lacks the functionality to accomplish the above goals. |
Beta Was this translation helpful? Give feedback.
-
For me, I'm just practicing training large language models. I wrote a version of alphaTex exporter on Python, starting from the PyGuitarPro GP reader. Though it's lacking some features like bar metadata and staves. I converted an archive of 47299 gp3-5 files to alphaTex, and created a dataset on HuggingFace. I'm currently training a couple of models: a GPT2 on the full guitartab dataset, and a GPT2 on just 4-string bass tracks. Longer term, I was thinking about coupling alphaTex with the corresponding audio data, and looking into training a model for auto-generating tabs from audio. Also, in more recent practical terms, I think @izszzz's idea of managing tabs under GitHub is amazing! |
Beta Was this translation helpful? Give feedback.
-
I couldn't find it in the code base, but is there an exporter into the AlphaTex format?
It would be amazing to have it, as it would be also a great input to train a language model, given large databases of existing GP files.
Beta Was this translation helpful? Give feedback.
All reactions