TTS-WebUI icon indicating copy to clipboard operation
TTS-WebUI copied to clipboard

add spotify's basic-pitch

Open rsxdalv opened this issue 11 months ago • 0 comments

https://github.com/spotify/basic-pitch

Basic Pitch is a Python library for Automatic Music Transcription (AMT), using lightweight neural network developed by Spotify's Audio Intelligence Lab. It's small, easy-to-use, pip install-able and npm install-able via its sibling repo.

The dependencies are mostly compatible; however -

"coremltools; platform_system == 'Darwin'",
"librosa>=0.8.0",
"mir_eval>=0.6",
"numpy>=1.18",
"**onnxruntime**; platform_system == 'Windows' and python_version < '3.11'",
"pretty_midi>=0.2.9",
"resampy>=0.2.2,<0.4.3",
"scikit-learn",
"scipy>=1.4.1",
"**tensorflow**>=2.4.1,<2.15.1; platform_system != 'Darwin' and python_version >= '3.11'",
"**tensorflow-macos**>=2.4.1,<2.15.1; platform_system == 'Darwin' and python_version > '3.11'",
"**tflite-runtime**; platform_system == 'Linux' and python_version < '3.11'",
"typing_extensions",

It might require tensorflow which is not yet integrated into the TTS WebUI

original issue: https://github.com/rsxdalv/tts-generation-webui/issues/434

rsxdalv avatar May 07 '25 20:05 rsxdalv