Update app.py

app.py

@@ -14,7 +14,7 @@ import re
 import time
 from streaming_stt_nemo import Model
 
-number_to_word = inflect.engine()
+# number_to_word = inflect.engine()
 
 default_lang = "en"
 
@@ -51,24 +51,24 @@ def model(text):
 
     return output
 
-def number_to_words(str):
-    words = str.split(' ')
-    result = []
-    
-    for word in words:
-        if( any(char.isdigit() for char in word) ):
-            word = number_to_word.number_to_words(word)
-
-        result.append(word)
-
-    final_result = ' '.join(result).replace('point', '')
-    return final_result
+#def number_to_words(str):
+#    words = str.split(' ')
+#    result = []
+#    
+#    for word in words:
+#        if( any(char.isdigit() for char in word) ):
+#            word = number_to_word.number_to_words(word)
+#
+#        result.append(word)
+#
+#    final_result = ' '.join(result).replace('point', '')
+#    return final_result
 
 async def respond(audio):
     user = transcribe(audio)
     reply = model(user)
-    reply2 = number_to_words(reply)
-
+#    reply2 = number_to_words(reply)
+#
     voices = await VoicesManager.create()
     voice = voices.find(Gender="Male", Locale="en-GB")
     
@@ -119,16 +119,18 @@ with gr.Blocks(css="style.css") as demo:
                      ### <center>Now Talk to - Dr. Watson</center>
                      """)
     with gr.Row():
-        user_input   = gr.Audio( label="Your Voice Chat", type="filepath", interactive=True,  elem_classes="microphone" )
-        output_audio = gr.Audio( label="WATSON",          type="filepath", interactive=False, elem_classes="audio", autoplay=True )
-    with gr.Row():
-        translate_btn = gr.Button("Response")
-        translate_btn.click( fn=respond, inputs=user_input, outputs=output_audio, api_name=False )
+        input = gr.Audio(label="Your Voice Chat", type="filepath", interactive=True, sources="microphone", waveform_options=False)
+        output = gr.Audio(label="WATSON", type="filepath", interactive=False, autoplay=True, elem_classes="audio")
+        gr.Interface(fn=respond, Inputs=[input], outputs=[output], live=True) 
+        
+    #with gr.Row():
+    #    translate_btn = gr.Button("Response")
+    #    translate_btn.click( fn=respond, inputs=user_input, outputs=output_audio, api_name=False )
     
     with gr.Row():
-        user_input   = gr.Textbox( label="Your Question", value="Dr. Watson can you summarize your adventures with Sherlock Holmes?")
-        input_text   = gr.Textbox( label="Input Text",    elem_id="important")
-        output_audio = gr.Audio( label="WATSON",          type="filepath", interactive=False, elem_classes="audio", autoplay=True )
+        user_input = gr.Textbox(label="Your Question", value="Dr. Watson can you summarize your adventures with Sherlock Holmes?")
+        input_text = gr.Textbox(label="Input Text", elem_id="important")
+        output_audio = gr.Audio(label="WATSON", type="filepath", interactive=False, autoplay=True, elem_classes="audio")
     with gr.Row():
         translate_btn = gr.Button("Response")
         translate_btn.click(fn=generate1, inputs=user_input, outputs=output_audio, api_name="translate")  
@@ -136,6 +138,6 @@ with gr.Blocks(css="style.css") as demo:
 
 
 if __name__ == "__main__":
-    demo.queue(max_size=200,api_open=False).launch()
+    demo.queue(max_size=200, api_open=False).launch()