Update src/pronunciation_checker.py

src/pronunciation_checker.py

@@ -35,15 +35,19 @@ class PronunciationChecker:
         return wav
 
 
-    def preprocess_wav(self, wav_path):
+    def preprocess_wav(self, wav_path, do_trim_silences=True):
         temp_audio_path = "temp.wav"
         audio_segment = process_wav(wav_path, 16000, do_trim_silences=False)
         audio_segment.export(temp_audio_path, format="wav")
 
-        wav = self.trim_silences_with_silero_vad(temp_audio_path)
+        if do_trim_silences:
+            wav = self.trim_silences_with_silero_vad(temp_audio_path)
+        else:
+            wav = self.read_audio(temp_audio_path)
 
         return wav, self.sr
 
+
     def extract_features(self, wav, layer=None):
         inputs = self.processor(wav.squeeze().to(self.device), sampling_rate=16000, return_tensors="pt", padding=True)
         inputs = {key: val.to(self.device) for key, val in inputs.items()}