debug

.idea/workspace.xml

@@ -6,6 +6,7 @@
   <component name="ChangeListManager">
     <list default="true" id="23565123-73ab-4f40-a9ef-1086e0c9e1ec" name="Changes" comment="">
       <change beforePath="$PROJECT_DIR$/.idea/workspace.xml" beforeDir="false" afterPath="$PROJECT_DIR$/.idea/workspace.xml" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/preprocess.py" beforeDir="false" afterPath="$PROJECT_DIR$/preprocess.py" afterDir="false" />
       <change beforePath="$PROJECT_DIR$/script.py" beforeDir="false" afterPath="$PROJECT_DIR$/script.py" afterDir="false" />
     </list>
     <option name="SHOW_DIALOG" value="false" />
@@ -99,7 +100,7 @@
       <updated>1742573353560</updated>
       <workItem from="1742573355153" duration="587000" />
       <workItem from="1742806974298" duration="2741000" />
-      <workItem from="1742810431420" duration="11063000" />
+      <workItem from="1742810431420" duration="12114000" />
     </task>
     <servers />
   </component>

preprocess.py

@@ -1,5 +1,6 @@
 
 import librosa
+import soundfile as sf
 import numpy as np
 import torch
 
@@ -16,7 +17,11 @@ def pad_audio(x, max_len=48000):
 def preprocess(audio_file):
     print(f'Preprocessing {audio_file}')
     # Load the audio file
-    y, sr = librosa.load(audio_file, sr=16000)
+    # y, sr = librosa.load(audio_file, sr=16000)
+
+    y, sr = sf.read(audio_file)
+    y = librosa.resample(y, orig_sr=sr, target_sr=16000)
+    sr = 16000
 
     # Evaluate N windows of the audio file
     num_eval = 5

script.py

@@ -94,7 +94,7 @@ for el in tqdm.tqdm(dataset_remote):
             # # OLD MODEL
             # score = model(tensor.to(device)).cpu().item()
             # RAWNNET2 MODEL
-            score = model(tensor.to(device))[:, 0].cpu()
+            score = model(tensor.to(device))[:, 1].cpu()
             print(f'SCORE OUT: {score}')
             score = score.mean().item()
             print(f'SCORE FINAL: {score}')