OGSD V0.1.1

This is a minor update to the initial release which only has these changes.

notes:

• corrected pronounciation on Label 23 ([f], [t], [p], [T], & [s] follow [xj] in semivowel transitions).
• added a missing phone from the included dictionary files (hotfix)

license still remains the same until the next major update.

OGSD Dataset with wav files

This is the initial release of the dataset with wav files included.

the audio included consists of around 40 mins of singing, and 5 minutes of rapping, and are sepperated by folders based on the vocal style.
stringed audio will be included at a later update.

the package can be downloaded from below under the Assets tab.
for more information on usage, please check the README of the repo.

Phoneme Distribution

MIDI Pitch Distribution