made some improve

main.py

@@ -0,0 +1,19 @@
+from transcribeHallu import loadModel
+from transcribeHallu import transcribePrompt
+
+##### The audio language may be different from the one for the output transcription.
+path = "data/KatyPerry-Firework.mp3"
+
+##### Activate this for music file to get a minimal processing
+isMusic = False
+
+##### Need to be adapted for each language.
+##### For prompt examples, see transcribeHallu.py getPrompt(lng:str)
+lng = "en"
+
+##### Model size to use
+modelSize = "medium"
+loadModel("0", modelSize=modelSize)
+
+result = transcribePrompt(path=path, lng=lng)
+print(result)