A lightweight yet powerful audio-to-MIDI converter with pitch bend detection
Python audiolightweightmachine-learningmidimusicpitch-detectionpolyphonicpythontranscriptiontypescript
change default value to 127.70ms that correspond to the expected default value of 11 frames
Vincent Degroote265a0687236b
diff --git a/basic_pitch/inference.py b/basic_pitch/inference.pyindex 715e019..2b6b3e1 100644--- a/basic_pitch/inference.py+++ b/basic_pitch/inference.py@@ -265,7 +265,7 @@ def predict(model_or_model_path: Union[keras.Model, pathlib.Path, str] = ICASSP_2022_MODEL_PATH,onset_threshold: float = 0.5,frame_threshold: float = 0.3,- minimum_note_length: float = 58,+ minimum_note_length: float = 127.70,minimum_frequency: Optional[float] = None,maximum_frequency: Optional[float] = None,multiple_pitch_bends: bool = False,