audio2midi 0.7.0__tar.gz → 0.9.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (16) hide show
  1. {audio2midi-0.7.0 → audio2midi-0.9.0}/PKG-INFO +8 -4
  2. {audio2midi-0.7.0 → audio2midi-0.9.0}/README.md +7 -3
  3. {audio2midi-0.7.0 → audio2midi-0.9.0}/pyproject.toml +1 -1
  4. {audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/basic_pitch_pitch_detector.py +2 -1
  5. {audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/magenta_music_transcription.py +1 -1
  6. {audio2midi-0.7.0 → audio2midi-0.9.0}/.gitignore +0 -0
  7. {audio2midi-0.7.0 → audio2midi-0.9.0}/.python-version +0 -0
  8. {audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/__init__.py +0 -0
  9. {audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/crepe_pitch_detector.py +0 -0
  10. {audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/crepe_pitch_detector_tf.py +0 -0
  11. {audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/librosa_pitch_detector.py +0 -0
  12. {audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/melodia_pitch_detector.py +0 -0
  13. {audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/mt3_music_transcription.py +0 -0
  14. {audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/pop2piano.py +0 -0
  15. {audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/py.typed +0 -0
  16. {audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/violin_pitch_detector.py +0 -0
{audio2midi-0.7.0 → audio2midi-0.9.0}/PKG-INFO RENAMED
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: audio2midi
3
- Version: 0.7.0
3
+ Version: 0.9.0
4
4
  Summary: Audio To Midi
5
5
  Author-email: dummyjenil <dummyjenil@gmail.com>
6
6
  Provides-Extra: all
@@ -97,7 +97,7 @@ Description-Content-Type: text/markdown
97
97
  ---
98
98
 
99
99
  ```bash
100
- pip install audio2midi[all] audio2midi[pop2piano] audio2midi[violin_pitch_detector] audio2midi[crepe_pitch_detector] audio2midi[crepe_pitch_detector_tf] audio2midi[melodia_pitch_detector] audio2midi[basic_pitch_pitch_detector] audio2midi[librosa_pitch_detector]
100
+ pip install audio2midi[all] audio2midi[pop2piano] audio2midi[violin_pitch_detector] audio2midi[crepe_pitch_detector] audio2midi[crepe_pitch_detector_tf] audio2midi[melodia_pitch_detector] audio2midi[basic_pitch_pitch_detector] audio2midi[librosa_pitch_detector] audio2midi[magenta_music_transcription]
101
101
  ```
102
102
  ---
103
103
 
@@ -146,6 +146,7 @@ from audio2midi.basic_pitch_pitch_detector import BasicPitch
146
146
  from audio2midi.crepe_pitch_detector import Crepe
147
147
  from audio2midi.violin_pitch_detector import Violin_Pitch_Det
148
148
  from audio2midi.pop2piano import Pop2Piano
149
+ from audio2midi.magenta_music_transcription import Magenta
149
150
  from torch import device as Device
150
151
  from torch.cuda import is_available as cuda_is_available
151
152
  device = Device("cuda" if cuda_is_available() else "cpu")
@@ -153,6 +154,7 @@ Crepe().predict(audio_path)
153
154
  Pop2Piano(device=device).predict(audio_path)
154
155
  Violin_Pitch_Det(device=device).predict(audio_path)
155
156
  BasicPitch(device=device).predict(audio_path)
157
+ Magenta().predict(audio_path)
156
158
  ```
157
159
 
158
160
  ---
@@ -166,6 +168,7 @@ from audio2midi.melodia_pitch_detector import Melodia
166
168
  from audio2midi.pop2piano import Pop2Piano
167
169
  from audio2midi.violin_pitch_detector import Violin_Pitch_Det
168
170
  from audio2midi.mt3_music_transcription import MT3
171
+ from audio2midi.magenta_music_transcription import Magenta
169
172
  from os import environ
170
173
  from huggingface_hub import hf_hub_download
171
174
  from shutil import unpack_archive
@@ -219,12 +222,13 @@ gr.TabbedInterface([
219
222
  gr.Interface(Normal_Pitch_Det().predict,[gr.Audio(type="filepath",label="Input Audio"),gr.Number(120,label="BPM"),gr.Number(512,label="HOP Len"),gr.Number(2,label="minimum note length"),gr.Number(0.1,label="threshold")],gr.File(label="Midi File")),
220
223
  gr.Interface(Guitar_Pitch_Det().predict,[gr.Audio(type="filepath",label="Input Audio"),gr.Number(4,label="mag_exp"),gr.Number(-61,label="Threshold"),gr.Number(6,label="Pre_post_max"),gr.Checkbox(False,label="backtrack"),gr.Checkbox(False,label="round_to_sixteenth"),gr.Number(1024,label="hop_length"),gr.Number(72,label="n_bins"),gr.Number(12,label="bins_per_octave")],gr.File(label="Midi File")),
221
224
  gr.Interface(Melodia().predict,[gr.Audio(type="filepath",label="Input Audio"),gr.Number(120,label="BPM",step=30),gr.Number(0.25,label="smoothness",step=0.05,info="Smooth the pitch sequence with a median filter of the provided duration (in seconds)."),gr.Number(0.1,label="minimum duration",step=0.1,info="Minimum allowed duration for note (in seconds). Shorter notes will be removed."),gr.Number(128,label="HOP")],gr.File(label="Midi File")),
222
- gr.Interface(BasicPitch(device=device).predict,[gr.Audio(type="filepath", label="Upload Audio"),gr.Number(0.5,label="onset_thresh",info="Minimum amplitude of an onset activation to be considered an onset."),gr.Number(0.3,label="frame_thresh",info="Minimum energy requirement for a frame to be considered present."),gr.Number(127.70,label="min_note_len",info="The minimum allowed note length in milliseconds."),gr.Number(120,label="midi_tempo"),gr.Checkbox(True,label="infer_onsets",info="add additional onsets when there are large differences in frame amplitudes."),gr.Checkbox(True,label="include_pitch_bends",info="include pitch bends."),gr.Checkbox(False,label="multiple_pitch_bends",info="allow overlapping notes in midi file to have pitch bends."),gr.Checkbox(True,label="melodia_trick",info="Use the melodia post-processing step.")],gr.File(label="Download Midi File")),
225
+ gr.Interface(BasicPitch(device=device).predict,[gr.Audio(type="filepath", label="Upload Audio"),gr.Number(0.5,label="onset_thresh",info="Minimum amplitude of an onset activation to be considered an onset."),gr.Number(0.3,label="frame_thresh",info="Minimum energy requirement for a frame to be considered present."),gr.Number(11,label="min_note_len",info="The minimum allowed note length in milliseconds."),gr.Number(120,label="midi_tempo"),gr.Checkbox(True,label="infer_onsets",info="add additional onsets when there are large differences in frame amplitudes."),gr.Checkbox(True,label="include_pitch_bends",info="include pitch bends."),gr.Checkbox(False,label="multiple_pitch_bends",info="allow overlapping notes in midi file to have pitch bends."),gr.Checkbox(True,label="melodia_trick",info="Use the melodia post-processing step.")],gr.File(label="Download Midi File")),
226
+ gr.Interface(Magenta().predict,[gr.Audio(type="filepath", label="Upload Audio"),gr.Number(0.5,label="onset_thresh",info="Minimum amplitude of an onset activation to be considered an onset."),gr.Number(3,label="min_note_len",info="The minimum allowed note length"),gr.Number(3,label="gap_tolerance_frames"),gr.Number(4,label="pitch_bend_steps"),gr.Number(1500,label="pitch_bend_depth"),gr.Checkbox(True,label="include_pitch_bends",info="include pitch bends.")],gr.File(label="Download Midi File")),
223
227
  gr.Interface(Violin_Pitch_Det(device=device,model_capacity=getenv("violin_model_capacity","full")).predict, [gr.Audio(label="Upload your Audio file",type="filepath"),gr.Number(32,label="Batch size"),gr.Radio(["spotify","tiktok"],value="spotify",label="Post Processing"),gr.Checkbox(True,label="include_pitch_bends")],gr.File(label="Download MIDI file")),
224
228
  gr.Interface(Crepe(getenv("crepe_model_capacity","full")).predict,[gr.Audio(type="filepath",label="Input Audio"),gr.Checkbox(False,label="viterbi",info="Apply viterbi smoothing to the estimated pitch curve"),gr.Checkbox(True,label="center"),gr.Number(10,label="step size",info="The step size in milliseconds for running pitch estimation."),gr.Number(0.8,label="minimum confidence"),gr.Number(32,label="batch size")],gr.File(label="Midi File")),
225
229
  gr.Interface(CrepeTF(getenv("crepe_model_capacity","full")).predict,[gr.Audio(type="filepath",label="Input Audio"),gr.Checkbox(False,label="viterbi",info="Apply viterbi smoothing to the estimated pitch curve"),gr.Checkbox(True,label="center"),gr.Number(10,label="step size",info="The step size in milliseconds for running pitch estimation."),gr.Number(0.8,label="minimum confidence"),gr.Number(32,label="batch size")],gr.File(label="Midi File")),
226
230
  gr.Interface(Pop2Piano(device).predict,[gr.Audio(label="Input Audio",type="filepath"),gr.Number(1, minimum=1, maximum=21, label="Composer"),gr.Number(2,label="Details in Piano"),gr.Number(1,label="Efficiency of Piano"),gr.Radio([1,2,4],label="steps per beat",value=2)],gr.File(label="MIDI File")),
227
231
  gr.Interface(MT3(str(Path("mt3_model").absolute())).predict,[gr.Audio(label="Input Audio",type="filepath"),gr.Number(0,label="seed")],gr.File(label="MIDI File")),
228
232
  midi_viz_ui
229
- ],["Normal Pitch Detection","Guitar Based Pitch Detection","Melodia","Spotify Pitch Detection","Violin Based Pitch Detection","Crepe Pitch Detection","Crepe Pitch Detection TF","Pop2Piano","MT3","Midi Vizulizer"]).launch()
233
+ ],["Normal Pitch Detection","Guitar Based Pitch Detection","Melodia","Spotify Pitch Detection","Magenta Pitch Detection","Violin Based Pitch Detection","Crepe Pitch Detection","Crepe Pitch Detection TF","Pop2Piano","MT3","Midi Vizulizer"]).launch()
230
234
  ```
{audio2midi-0.7.0 → audio2midi-0.9.0}/README.md RENAMED
@@ -5,7 +5,7 @@
5
5
  ---
6
6
 
7
7
  ```bash
8
- pip install audio2midi[all] audio2midi[pop2piano] audio2midi[violin_pitch_detector] audio2midi[crepe_pitch_detector] audio2midi[crepe_pitch_detector_tf] audio2midi[melodia_pitch_detector] audio2midi[basic_pitch_pitch_detector] audio2midi[librosa_pitch_detector]
8
+ pip install audio2midi[all] audio2midi[pop2piano] audio2midi[violin_pitch_detector] audio2midi[crepe_pitch_detector] audio2midi[crepe_pitch_detector_tf] audio2midi[melodia_pitch_detector] audio2midi[basic_pitch_pitch_detector] audio2midi[librosa_pitch_detector] audio2midi[magenta_music_transcription]
9
9
  ```
10
10
  ---
11
11
 
@@ -54,6 +54,7 @@ from audio2midi.basic_pitch_pitch_detector import BasicPitch
54
54
  from audio2midi.crepe_pitch_detector import Crepe
55
55
  from audio2midi.violin_pitch_detector import Violin_Pitch_Det
56
56
  from audio2midi.pop2piano import Pop2Piano
57
+ from audio2midi.magenta_music_transcription import Magenta
57
58
  from torch import device as Device
58
59
  from torch.cuda import is_available as cuda_is_available
59
60
  device = Device("cuda" if cuda_is_available() else "cpu")
@@ -61,6 +62,7 @@ Crepe().predict(audio_path)
61
62
  Pop2Piano(device=device).predict(audio_path)
62
63
  Violin_Pitch_Det(device=device).predict(audio_path)
63
64
  BasicPitch(device=device).predict(audio_path)
65
+ Magenta().predict(audio_path)
64
66
  ```
65
67
 
66
68
  ---
@@ -74,6 +76,7 @@ from audio2midi.melodia_pitch_detector import Melodia
74
76
  from audio2midi.pop2piano import Pop2Piano
75
77
  from audio2midi.violin_pitch_detector import Violin_Pitch_Det
76
78
  from audio2midi.mt3_music_transcription import MT3
79
+ from audio2midi.magenta_music_transcription import Magenta
77
80
  from os import environ
78
81
  from huggingface_hub import hf_hub_download
79
82
  from shutil import unpack_archive
@@ -127,12 +130,13 @@ gr.TabbedInterface([
127
130
  gr.Interface(Normal_Pitch_Det().predict,[gr.Audio(type="filepath",label="Input Audio"),gr.Number(120,label="BPM"),gr.Number(512,label="HOP Len"),gr.Number(2,label="minimum note length"),gr.Number(0.1,label="threshold")],gr.File(label="Midi File")),
128
131
  gr.Interface(Guitar_Pitch_Det().predict,[gr.Audio(type="filepath",label="Input Audio"),gr.Number(4,label="mag_exp"),gr.Number(-61,label="Threshold"),gr.Number(6,label="Pre_post_max"),gr.Checkbox(False,label="backtrack"),gr.Checkbox(False,label="round_to_sixteenth"),gr.Number(1024,label="hop_length"),gr.Number(72,label="n_bins"),gr.Number(12,label="bins_per_octave")],gr.File(label="Midi File")),
129
132
  gr.Interface(Melodia().predict,[gr.Audio(type="filepath",label="Input Audio"),gr.Number(120,label="BPM",step=30),gr.Number(0.25,label="smoothness",step=0.05,info="Smooth the pitch sequence with a median filter of the provided duration (in seconds)."),gr.Number(0.1,label="minimum duration",step=0.1,info="Minimum allowed duration for note (in seconds). Shorter notes will be removed."),gr.Number(128,label="HOP")],gr.File(label="Midi File")),
130
- gr.Interface(BasicPitch(device=device).predict,[gr.Audio(type="filepath", label="Upload Audio"),gr.Number(0.5,label="onset_thresh",info="Minimum amplitude of an onset activation to be considered an onset."),gr.Number(0.3,label="frame_thresh",info="Minimum energy requirement for a frame to be considered present."),gr.Number(127.70,label="min_note_len",info="The minimum allowed note length in milliseconds."),gr.Number(120,label="midi_tempo"),gr.Checkbox(True,label="infer_onsets",info="add additional onsets when there are large differences in frame amplitudes."),gr.Checkbox(True,label="include_pitch_bends",info="include pitch bends."),gr.Checkbox(False,label="multiple_pitch_bends",info="allow overlapping notes in midi file to have pitch bends."),gr.Checkbox(True,label="melodia_trick",info="Use the melodia post-processing step.")],gr.File(label="Download Midi File")),
133
+ gr.Interface(BasicPitch(device=device).predict,[gr.Audio(type="filepath", label="Upload Audio"),gr.Number(0.5,label="onset_thresh",info="Minimum amplitude of an onset activation to be considered an onset."),gr.Number(0.3,label="frame_thresh",info="Minimum energy requirement for a frame to be considered present."),gr.Number(11,label="min_note_len",info="The minimum allowed note length in milliseconds."),gr.Number(120,label="midi_tempo"),gr.Checkbox(True,label="infer_onsets",info="add additional onsets when there are large differences in frame amplitudes."),gr.Checkbox(True,label="include_pitch_bends",info="include pitch bends."),gr.Checkbox(False,label="multiple_pitch_bends",info="allow overlapping notes in midi file to have pitch bends."),gr.Checkbox(True,label="melodia_trick",info="Use the melodia post-processing step.")],gr.File(label="Download Midi File")),
134
+ gr.Interface(Magenta().predict,[gr.Audio(type="filepath", label="Upload Audio"),gr.Number(0.5,label="onset_thresh",info="Minimum amplitude of an onset activation to be considered an onset."),gr.Number(3,label="min_note_len",info="The minimum allowed note length"),gr.Number(3,label="gap_tolerance_frames"),gr.Number(4,label="pitch_bend_steps"),gr.Number(1500,label="pitch_bend_depth"),gr.Checkbox(True,label="include_pitch_bends",info="include pitch bends.")],gr.File(label="Download Midi File")),
131
135
  gr.Interface(Violin_Pitch_Det(device=device,model_capacity=getenv("violin_model_capacity","full")).predict, [gr.Audio(label="Upload your Audio file",type="filepath"),gr.Number(32,label="Batch size"),gr.Radio(["spotify","tiktok"],value="spotify",label="Post Processing"),gr.Checkbox(True,label="include_pitch_bends")],gr.File(label="Download MIDI file")),
132
136
  gr.Interface(Crepe(getenv("crepe_model_capacity","full")).predict,[gr.Audio(type="filepath",label="Input Audio"),gr.Checkbox(False,label="viterbi",info="Apply viterbi smoothing to the estimated pitch curve"),gr.Checkbox(True,label="center"),gr.Number(10,label="step size",info="The step size in milliseconds for running pitch estimation."),gr.Number(0.8,label="minimum confidence"),gr.Number(32,label="batch size")],gr.File(label="Midi File")),
133
137
  gr.Interface(CrepeTF(getenv("crepe_model_capacity","full")).predict,[gr.Audio(type="filepath",label="Input Audio"),gr.Checkbox(False,label="viterbi",info="Apply viterbi smoothing to the estimated pitch curve"),gr.Checkbox(True,label="center"),gr.Number(10,label="step size",info="The step size in milliseconds for running pitch estimation."),gr.Number(0.8,label="minimum confidence"),gr.Number(32,label="batch size")],gr.File(label="Midi File")),
134
138
  gr.Interface(Pop2Piano(device).predict,[gr.Audio(label="Input Audio",type="filepath"),gr.Number(1, minimum=1, maximum=21, label="Composer"),gr.Number(2,label="Details in Piano"),gr.Number(1,label="Efficiency of Piano"),gr.Radio([1,2,4],label="steps per beat",value=2)],gr.File(label="MIDI File")),
135
139
  gr.Interface(MT3(str(Path("mt3_model").absolute())).predict,[gr.Audio(label="Input Audio",type="filepath"),gr.Number(0,label="seed")],gr.File(label="MIDI File")),
136
140
  midi_viz_ui
137
- ],["Normal Pitch Detection","Guitar Based Pitch Detection","Melodia","Spotify Pitch Detection","Violin Based Pitch Detection","Crepe Pitch Detection","Crepe Pitch Detection TF","Pop2Piano","MT3","Midi Vizulizer"]).launch()
141
+ ],["Normal Pitch Detection","Guitar Based Pitch Detection","Melodia","Spotify Pitch Detection","Magenta Pitch Detection","Violin Based Pitch Detection","Crepe Pitch Detection","Crepe Pitch Detection TF","Pop2Piano","MT3","Midi Vizulizer"]).launch()
138
142
  ```
{audio2midi-0.7.0 → audio2midi-0.9.0}/pyproject.toml RENAMED
@@ -1,6 +1,6 @@
1
1
  [project]
2
2
  name = "audio2midi"
3
- version = "0.7.0"
3
+ version = "0.9.0"
4
4
  description = "Audio To Midi"
5
5
  readme = "README.md"
6
6
  authors = [
{audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/basic_pitch_pitch_detector.py RENAMED
@@ -778,6 +778,7 @@ class BasicPitch():
778
778
  }
779
779
  return unwrapped_output
780
780
 
781
- def predict(self,audio,onset_thresh=0.5,frame_thresh=0.3,min_note_len=127.70,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"):
781
+ 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"):
782
+ int(np.round(min_note_len / 1000 * (AUDIO_SAMPLE_RATE / FFT_HOP)))
782
783
  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)
783
784
  return output_file
{audio2midi-0.7.0 → audio2midi-0.9.0}/src/audio2midi/magenta_music_transcription.py RENAMED
@@ -71,6 +71,6 @@ class Magenta:
71
71
  return actProb , onProb , offProb , volProb , tempo
72
72
 
73
73
 
74
- def predict(self,audio,onset_thresh=0,min_note_len=3,include_pitch_bends=True,gap_tolerance_frames = 3,pitch_bend_depth = 1500,pitch_bend_steps = 4,progress_callback: Callable[[int, int], None] = None,output_file="output.mid"):
74
+ def predict(self,audio,onset_thresh=0,min_note_len=3,gap_tolerance_frames = 3,pitch_bend_depth = 1500,pitch_bend_steps = 4,include_pitch_bends=True,progress_callback: Callable[[int, int], None] = None,output_file="output.mid"):
75
75
  model_output_to_notes(self.run_inference(audio,progress_callback),onset_thresh = onset_thresh,min_note_len = min_note_len,include_pitch_bends = include_pitch_bends,pitch_bend_depth=pitch_bend_depth,pitch_bend_steps=pitch_bend_steps,gap_tolerance_frames=gap_tolerance_frames).write(output_file)
76
76
  return output_file
{audio2midi-0.7.0 → audio2midi-0.9.0}/.gitignore RENAMED
File without changes
{audio2midi-0.7.0 → audio2midi-0.9.0}/.python-version RENAMED
File without changes