Upload 3 files

app.py

@@ -3,14 +3,23 @@
 # %% auto 0
 __all__ = ['learn', 'categories', 'aud', 'examples', 'intf', 'log_mel_spec_tfm', 'classify_aud']
 
+from fastai.vision.all import *
+import librosa.display
+import matplotlib.pyplot as plt
+import numpy as np
+from functools import partial
+import pandas as pd
+import librosa
+from scipy.io import wavfile
+import gradio as gr 
+
 # %% app.ipynb 1
-def log_mel_spec_tfm(fname, src_path, dst_path):
-    os.makedirs(str(dst_path), exist_ok = True)
-    y, sr = librosa.load(str(src_path/fname), mono=True)
+def log_mel_spec_tfm(fname):
+    y, sr = librosa.load(fname, mono=True)
     D = librosa.amplitude_to_db(np.abs(librosa.stft(y)), ref=np.max)
     img = librosa.display.specshow(D, y_axis='linear', x_axis='time',
                                   sr=sr)
-    plt.savefig(str(dst_path/fname[:-4]) + '.png')
+    plt.savefig(str(fname[:-4]) + '.png')
     plt.close()
     return img
 
@@ -18,18 +27,18 @@ def log_mel_spec_tfm(fname, src_path, dst_path):
 learn = load_learner('model.pkl')
 learn.remove_cb(ProgressCallback)
 
-# %% app.ipynb 6
+# %% app.ipynb 5
 categories = ('Brass', 'Flute', 'Guitar', 'Keyboard', 'Mallet', 'Reed', 'String', 'Vocal')
 
 def classify_aud(aud):
-    log_mel_spec_tfm(aud, Path('.'), Path('.'))
+    log_mel_spec_tfm(aud)
     img_fname = str(aud[:-4]) + '.png'
     pred, idx, probs = learn.predict(img_fname)
     return dict(zip(categories, map(float, probs)))
 
-# %% app.ipynb 8
+# %% app.ipynb 6
 aud = gr.Audio(source="upload", type="numpy")
-examples = ['test/' + str(f.name) for f in Path('test').iterdir()]
+examples = [f.name for f in Path('.').iterdir() if '.wav' in f.name]
 
 intf = gr.Interface(fn = classify_aud, inputs = aud, outputs = "label", examples = examples)
 intf.launch(inline = False)

requirements.txt

@@ -1,11 +1 @@
-fastbook
-fastai
-librosa
-matplotlib
-numpy
-functools
-pathlib
-multiprocessing
-os
-scipy
-gradio
+