audio2midi 0.9.0__tar.gz → 0.11.0__tar.gz

{audio2midi-0.9.0 → audio2midi-0.11.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: audio2midi
-Version: 0.9.0
+Version: 0.11.0
 Summary: Audio To Midi
 Author-email: dummyjenil <dummyjenil@gmail.com>
 Provides-Extra: all

{audio2midi-0.9.0 → audio2midi-0.11.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "audio2midi"
-version = "0.9.0"
+version = "0.11.0"
 description = "Audio To Midi"
 readme = "README.md"
 authors = [

{audio2midi-0.9.0 → audio2midi-0.11.0}/src/audio2midi/basic_pitch_pitch_detector.py

@@ -778,7 +778,6 @@ class BasicPitch():
         }
         return unwrapped_output
 
-    def predict(self,audio,onset_thresh=0.5,frame_thresh=0.3,min_note_len=11,midi_tempo=120,infer_onsets=True,include_pitch_bends=True,multiple_pitch_bends=True,melodia_trick=True,progress_callback: Callable[[int, int], None] = None,min_freqat=None,max_freqat=None,output_file="output.mid"):
-        int(np.round(min_note_len / 1000 * (AUDIO_SAMPLE_RATE / FFT_HOP)))
+    def predict(self,audio,onset_thresh=0.5,frame_thresh=0.3,min_note_len=11,midi_tempo=120,infer_onsets=True,include_pitch_bends=True,multiple_pitch_bends=False,melodia_trick=True,progress_callback: Callable[[int, int], None] = None,min_freqat=None,max_freqat=None,output_file="output.mid"):
         model_output_to_notes(self.run_inference(audio,progress_callback),onset_thresh  = onset_thresh,frame_thresh  = frame_thresh,infer_onsets  = infer_onsets,min_note_len  = min_note_len,min_freq  = min_freqat,max_freq  = max_freqat,include_pitch_bends  = include_pitch_bends,multiple_pitch_bends  = multiple_pitch_bends,melodia_trick  = melodia_trick,midi_tempo  = midi_tempo).write(output_file)
         return output_file

{audio2midi-0.9.0 → audio2midi-0.11.0}/src/audio2midi/crepe_pitch_detector.py

@@ -913,7 +913,8 @@ class Crepe():
                 inputs = batch[0].to(device)
                 outputs = self.model(inputs)
                 all_outputs.append(outputs.cpu())
-                progress_callback(i,total_batch)
+                if progress_callback:
+                    progress_callback(i,total_batch)
         return torch.cat(all_outputs, dim=0)
 
     def model_predict(self,audio:np.ndarray,viterbi, center, step_size,progress_callback,batch_size):