language list and prompt engineering ;-)

app.py

@@ -174,7 +174,9 @@ iface = gr.Interface(
         gr.Image(type="pil", label="Camera Input"),
         gr.File(label="File Upload"),
         gr.Audio(sources="microphone", type="filepath", label="Mic Input"),
-        gr.Textbox(lines=2, label="Text Input")
+        gr.Textbox(lines=2, label="Text Input"),
+        gr.Dropdown(choices=TEXT_SOURCE_LANGUAGE_NAMES, label="Input Language"),
+        gr.Dropdown(choices=TEXT_SOURCE_LANGUAGE_NAMES, label="Target Language")
     ],
     outputs=[
         RichTextbox(label="Processed Text"),
@@ -222,34 +224,34 @@ if __name__ == "__main__":
 
 
 
-iface = gr.Interface(
-    fn=process_input,
-    inputs=[
-        gr.Image(type="pil", label="Camera Input"),
-        gr.File(label="File Upload"),
-        gr.Audio(sources="microphone", type="filepath", label="Mic Input"),
-        gr.Textbox(lines=2, label="Text Input"),
-        gr.Dropdown(choices=TEXT_SOURCE_LANGUAGE_NAMES, label="Input Language"),
-        gr.Dropdown(choices=TEXT_SOURCE_LANGUAGE_NAMES, label="Target Language")
-    ],
-    outputs=[
-        gr.RichTextbox(label="Processed Text"),
-        gr.Audio(label="Audio Output")
-    ],
-    title="OCR and Speech Processing App",
-    description="This app processes images, PDFs, and audio inputs to generate text and audio outputs."
-)
+# iface = gr.Interface(
+#     fn=process_input,
+#     inputs=[
+#         gr.Image(type="pil", label="Camera Input"),
+#         gr.File(label="File Upload"),
+#         gr.Audio(sources="microphone", type="filepath", label="Mic Input"),
+#         gr.Textbox(lines=2, label="Text Input"),
+#         gr.Dropdown(choices=TEXT_SOURCE_LANGUAGE_NAMES, label="Input Language"),
+#         gr.Dropdown(choices=TEXT_SOURCE_LANGUAGE_NAMES, label="Target Language")
+#     ],
+#     outputs=[
+#         gr.RichTextbox(label="Processed Text"),
+#         gr.Audio(label="Audio Output")
+#     ],
+#     title="OCR and Speech Processing App",
+#     description="This app processes images, PDFs, and audio inputs to generate text and audio outputs."
+# )
 
-if __name__ == "__main__":
-    iface.launch()
+# if __name__ == "__main__":
+# #     iface.launch()
 
-demo = gr.Interface(
-    lambda x:x,
-    RichTextbox(),  # interactive version of your component
-    RichTextbox(),  # static version of your component
-    examples=[[example]],  # uncomment this line to view the "example version" of your component
-)
+# demo = gr.Interface(
+#     lambda x:x,
+#     RichTextbox(),  # interactive version of your component
+#     RichTextbox(),  # static version of your component
+#     examples=[[example]],  # uncomment this line to view the "example version" of your component
+# )
 
 
-if __name__ == "__main__":
-    demo.launch()
+# if __name__ == "__main__":
+#     demo.launch()