41727469634d6f6e6b65793a32302e30332e32343a31393532

audio2text/a2t.py

@@ -6,9 +6,10 @@ LIMIT = 90 # limit 90 seconds
 
 class A2T:
     def __init__(self, mic):
-        self.mic = mic
+        print(type(mic.export().read()))
+        self.mic = mic.export().read()
 
-    def __preproccess(self, audio, frame_rate):
+    def __preprocces(self, audio, frame_rate):
         try:
             audio = audio / 32678.0
 
@@ -23,7 +24,7 @@ class A2T:
             audio = torch.tensor(audio)
             return audio
         except Exception as e:
-            print("Error", e)
+            print("Preprocces error", e)
             return None
         
     def predict(self):
@@ -35,12 +36,12 @@ class A2T:
     
         try:
             forced_decoder_ids = processor.get_decoder_prompt_ids(language="english", task="transcribe")
-            audio = self.__preproccess(audio=audio, frame_rate=frame_rate)
+            audio = self.__preprocces(audio=audio, frame_rate=frame_rate)
             inputs = processor(audio=audio, sampling_rate=16000, return_tensors="pt")
             predicted_ids = model.generate(**inputs, max_length=400, forced_decoder_ids=forced_decoder_ids)
             transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
             return transcription[0]
         except Exception as e:
-            print("Error", e)
+            print("Predict error", e)
             return "Oops some kinda error"