Refactor audio input and output

app.py

@@ -13,7 +13,7 @@ examples = examples_atco2 #+ examples_atcosim
 
 whisper = pipeline(model='jlvdoorn/whisper-large-v3-atco2-asr-atcosim')
 
-def transcribe(audio, model_version):
+def transcribe(audio):
     if audio is not None:
         return whisper(audio)['text']
     else:
@@ -21,12 +21,9 @@ def transcribe(audio, model_version):
 
 file_iface = gr.Interface(
     fn = transcribe,
-    inputs = [gr.Audio(source='upload', interactive=True),
-              gr.Checkbox(label='Transcribe only', default=False),
-              gr.Dropdown(choices=['large-v2', 'large-v3'], value='large-v3', label='Whisper model version')
-            ],
+    inputs = gr.Audio(source='upload', interactive=True),
 
-    outputs = [gr.Textbox(label='Transcription'), gr.Textbox(label='Callsigns, commands and values')],
+    outputs = gr.Textbox(label='Transcription'),
     title = 'Whisper ATC - Large v3',
     description = 'Transcribe ATC speech',
     # examples = examples,
@@ -34,12 +31,9 @@ file_iface = gr.Interface(
 
 mic_iface = gr.Interface(
     fn = transcribe,
-    inputs = [gr.Audio(source='microphone', type='filepath'),
-              gr.Checkbox(label='Transcribe only', default=False),
-              gr.Dropdown(choices=['large-v2', 'large-v3'], value='large-v3', label='Whisper model version')
-            ],
+    inputs = gr.Audio(source='microphone', type='filepath'),
 
-    outputs = [gr.Textbox(label='Transcription'), gr.Textbox(label='Callsigns, commands and values')],
+    outputs = gr.Textbox(label='Transcription'),
     title = 'Whisper ATC - Large v3',
     description = 'Transcribe ATC speech',
 )