Source Link Not Working

[kevinlindong]

Currently, the link to the source of 4.9 GB dataset is not working. Do you happen to know if there's anything I can do to download this dataset? Thanks for the help!

[nathangur]

I also cannot access the source dataset.

[curzel-it]

Hello! Just checking if there's any chance of getting access the original dataset...?

[wayne391]

Hi,

I personally don’t have a backup anymore...

I originally stored it on my educational Google Drive, but a few years after I graduated, the account was suddenly shut down. Unfortunately, I don’t have a copy of it anymore. I had spent a lot of time on it...
(Lesson learned: never trust any educational/free accounts, even from big tech)

---

Here are a few alternatives you might consider:

1. Re-crawling the data using my code
   You could try re-running my crawler. I know it was still working in some industry labs as recently as 2023.

2. Contacting my former lab and advisor
   You may reach out to my former advisor, Dr. Yi-Hsuan (Eric) Yang: https://affige.github.io/
   They are very kind people : )
   If I recall correctly, there may still be a copy stored on the lab’s NAS.

3. Sheetsage [highly recommend]
   Check out the Sheetsage (2022) project!
   It's a lead sheet transcription model based on the same source, HookTheory.
   To the best of my knowledge, it is one of the most up-to-date ones in the open source community.
   Here is the data handling repo of sheetsage.

[curzel-it]

Thank you for the tips! I’ll try out Sage Have a nice day!

…

On Thu, 2 Apr 2026 at 08:01, Wen-Yi (Wayne) Hsiao ***@***.***> wrote: *wayne391* left a comment (wayne391/lead-sheet-dataset#6) <#6 (comment)> Hi, I personally don’t have a backup anymore... I originally stored it on my educational Google Drive, but a few years after I graduated, the account was suddenly shut down. Unfortunately, I don’t have a copy of it anymore. I had spent a lot of time on it... (Lesson learned: never trust any educational/free accounts, even from big tech) ------------------------------ Here are a few alternatives you might consider: 1. Re-crawling the data using my code You could try re-running my crawler. I know it was still working in some industry labs as recently as 2023. 2. Contacting my former lab and advisor You may reach out to my former advisor, Dr. Yi-Hsuan (Eric) Yang: https://affige.github.io/ They are very kind people : ) If I recall correctly, there may still be a copy stored on the lab’s NAS. 3. Sheetsage [highly recommend] Check out the Sheetsage <https://github.com/chrisdonahue/sheetsage>(2022) project! It's a lead sheet transcription model based on the same source, HookTheory. To the best of my knowledge, it is one of the most up-to-date ones in the open source community. Here is the data handling repo <https://github.com/chrisdonahue/sheetsage-data> of sheetsage. — Reply to this email directly, view it on GitHub <#6?email_source=notifications&email_token=AALBHBW4FKIO7MXEGONPEYL4TX625A5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTIMJXGQ4DOMJRGE3KM4TFMFZW63VHMNXW23LFNZ2KKZLWMVXHJLDGN5XXIZLSL5RWY2LDNM#issuecomment-4174871116>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AALBHBUTGWCL7NX6JASSVOD4TX625AVCNFSM6AAAAACXKC2IDGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DCNZUHA3TCMJRGY> . You are receiving this because you commented.Message ID: ***@***.***>