Update asr.py

asr.py

@@ -67,19 +67,9 @@ model = Wav2Vec2ForCTC.from_pretrained(MODEL_ID)
 
 
 def transcribe(
-    audio_source=None, microphone=None, file_upload=None, lang="eng (English)"
-):
-    if type(microphone) is dict:
-        # HACK: microphone variable is a dict when running on examples
-        microphone = microphone["name"]
-    audio_fp = (
-        file_upload if "upload" in str(audio_source or "").lower() else microphone
-    )
-    
-    if audio_fp is None:
-        return "ERROR: You have to either use the microphone or upload an audio file"
-    
-    audio_samples = librosa.load(audio_fp, sr=ASR_SAMPLING_RATE, mono=True)[0]
+    audio_data, lang="eng (English)"
+):    
+    sr, audio_samples = audio_data
 
     lang_code = lang.split()[0]
     processor.tokenizer.set_target_lang(lang_code)