Update app.py

app.py

@@ -74,6 +74,7 @@ def audioMode(audio):
     print(type(audio))
     print(audio)
     text = pipe(audio, batch_size=8, generate_kwargs={"task": "transcribe"}, return_timestamps=True)["text"]
+    pred_text = textMode(text, 50)
     #sampling_rate = audio[0]
     #audio_array = audio[1]
     #print(sampling_rate)
@@ -81,7 +82,7 @@ def audioMode(audio):
     #input_features = processor_audio(audio_array, sampling_rate=16000, return_tensors="pt").input_features 
     #predicted_ids = model_audio.generate(input_features)
     #transcription = processor_audio.batch_decode(predicted_ids, skip_special_tokens=True)
-    return text
+    return pred_text
 
 
 interface_title = "TSAI-ERA-V1 - Capstone - Multimodal GPT Demo"