Add nemo inference code

app.py

@@ -25,20 +25,20 @@ def process_audio_file(file):
     return data
 
 
-def transcribe(file_mic, file_upload):
+def transcribe(Microphone, File_Upload):
     warn_output = ""
-    if (file_mic is not None) and (file_upload is not None):
+    if (Microphone is not None) and (File_Upload is not None):
         warn_output = "WARNING: You've uploaded an audio file and used the microphone. " \
                       "The recorded file from the microphone will be used and the uploaded audio will be discarded.\n"
-        file = file_mic
+        file = Microphone
 
-    elif (file_mic is None) and (file_upload is None):
+    elif (Microphone is None) and (File_Upload is None):
         return "ERROR: You have to either use the microphone or upload an audio file"
 
-    elif file_mic is not None:
-        file = file_mic
+    elif Microphone is not None:
+        file = Microphone
     else:
-        file = file_upload
+        file = File_Upload
 
     audio_data = process_audio_file(file)
 
@@ -47,6 +47,7 @@ def transcribe(file_mic, file_upload):
         soundfile.write(audio_path, audio_data, SAMPLE_RATE)
 
         transcriptions = model.transcribe([audio_path])
+
         # if transcriptions form a tuple (from RNNT), extract just "best" hypothesis
         if type(transcriptions) == tuple and len(transcriptions) == 2:
             transcriptions = transcriptions[0]
@@ -63,8 +64,8 @@ iface = gr.Interface(
     outputs="text",
     layout="horizontal",
     theme="huggingface",
-    title="NeMo Conformer Transducer Large",
-    description="Demo for speech recognition using Conformers",
+    title="NeMo Conformer Transducer Large - English",
+    description="Demo for English speech recognition using Conformer Transducers",
     enable_queue=True,
     allow_flagging=False,
 )