/LabelMakr

A GUI Toolkit for SVS Label Generation. Heavily utilizes SOFA & Whisper to generate htk-style force-aligned labels with a focus on singing.

Primary LanguagePythonMIT LicenseMIT

LabelMakr 🛋️

Screenshot of LabelMakr and the transcription editor.

LabelMakr is a GUI tool to help users easily generate SVS phoneme-level labels. It is intended for use with DiffSinger, but is easily adaptable for other systems. Currently, LabelMakr has full support for English, Japanese, Chinese, French and Korean singing!

Please use the portable version for Windows found here.

Community Contributions 🧑‍🤝‍🧑

Custom SOFA Model Implementation

Please check out the guide on how to implement custom SOFA models here!

Credits

Please check out credits here!

Manual Installation 🧰

Read the guide on how to manually install LabelMakr here!