Update app.py

app.py

@@ -120,6 +120,43 @@ def process_speech(input_language, audio_input):
     except Exception as e :
         return f"{e}"
 
+def translate_text(input_text, input_language):
+    """
+    Translate text from one language to another.
+    """
+    try:
+        text_translation_result = client.predict(
+            task="T2TT (Text to Text translation)",
+            input_text=input_text,
+            source_language=Auto-detect,  # Auto-detect or specified language
+            target_language=input_language,
+            api_name="/run_text_translation"  # Hypothetical API endpoint for text translation
+        )
+
+        translated_text = text_translation_result['translated_text']
+        return translated_text
+    except Exception as e:
+        return f"An error occurred during translation: {e}"
+
+def convert_text_to_speech(input_text, input_language):
+    """
+    Convert text to speech in the specified language.
+    """
+    try:
+        text_to_speech_result = client.predict(
+            task="T2ST (Text to Speech translation)",
+            input_text=input_text,
+            language=input_language,
+            api_name="/run_text_to_speech"  # Hypothetical API endpoint for text-to-speech
+        )
+
+        # Assuming the API returns a file path or similar identifier for the audio
+        audio_file = text_to_speech_result['audio_file']
+        return audio_file
+    except Exception as e:
+        return f"An error occurred during text-to-speech conversion: {e}"
+
+
 def save_image(image_input, output_dir="saved_images"):
     if not os.path.exists(output_dir):
         os.makedirs(output_dir)
@@ -376,6 +413,10 @@ def process_and_query(input_language=None, audio_input=None, image_input=None, t
         sources_info = vectara_response.get('sources', [])
 
 
+        # Convert translated text to speech
+        audio_file_path = convert_text_to_speech(translated_response, input_language)
+
+
         # Format Vectara response in Markdown
         markdown_output = "### Vectara Response Summary\n"
         markdown_output += f"* **Summary**: {summary}\n"
@@ -383,9 +424,15 @@ def process_and_query(input_language=None, audio_input=None, image_input=None, t
         for source in sources_info:
             markdown_output += f"* {source}\n"
 
-        # Process the summary with OpenAI
+        # Process the summary with Stablemed
         final_response = process_summary_with_stablemed(summary)
 
+        # Translate the final response
+        translated_response = translate_text(final_response, input_language)
+
+        # Convert translated text to speech
+        audio_file_path = convert_text_to_speech(translated_response, input_language)
+        
         # Evaluate hallucination
         hallucination_label = evaluate_hallucination(final_response, summary)
 
@@ -395,10 +442,10 @@ def process_and_query(input_language=None, audio_input=None, image_input=None, t
         markdown_output += "\n### Hallucination Evaluation\n"
         markdown_output += f"* **Label**: {hallucination_label}\n"
 
-        return markdown_output
+        return markdown_output, audio_file_path
         
     except Exception as e:
-        return f"Error occurred during processing: {e}. No hallucination evaluation."
+        return f"Error occurred during processing: {e}. No hallucination evaluation.", None
 
 
 
@@ -524,9 +571,10 @@ languages = [
     "Zulu"
 ]
 
+
 def clear():
-    # Return default values 
-    return "English", None, None, "", "", "", ""
+    # Return default values for each component
+    return "English", None, None, "", None
 
 
 
@@ -550,23 +598,26 @@ def create_interface():
 
         with gr.Accordion("MultiMed", open=False) as multimend_accordion:
             text_input = gr.Textbox(label="Use Text", lines=3, placeholder="I have had a sore throat and phlegm for a few days and now my cough has gotten worse!")
-            text_output = gr.Markdown(label="Output text")  # Markdown component for text
+        
+            gr.Examples([
+                ["What is the proper treatment for buccal herpes?"],
+                ["I have had a sore throat and hoarse voice for serveral days and now a strong cough recently "],
+                ["How does cellular metabolism work TCA cycle"],
+                ["What special care must be provided to children with chicken pox?"],
+                ["When and how often should I wash my hands?"],
+                ["بکل ہرپس کا صحیح علاج کیا ہے؟"],
+                ["구강 헤르페스의 적절한 치료법은 무엇입니까?"],
+                ["Je, ni matibabu gani sahihi kwa herpes ya buccal?"],
+            ],inputs=[text_input])
+        text_output = gr.Markdown(label="MultiMed")  # Markdown component for text
+        audio_output = gr.Audio(label="Audio Out", type="filepath")
 
-        text_button = gr.Button("Use MultiMed")
-        text_button.click(process_and_query, inputs=[input_language, audio_input, image_input, text_input], outputs=[text_output])
-        gr.Examples([
-            ["What is the proper treatment for buccal herpes?"],
-            ["Male, 40 presenting with swollen glands and a rash"],
-            ["How does cellular metabolism work TCA cycle"],
-            ["What special care must be provided to children with chicken pox?"],
-            ["When and how often should I wash my hands?"],
-            ["بکل ہرپس کا صحیح علاج کیا ہے؟"],
-            ["구강 헤르페스의 적절한 치료법은 무엇입니까?"],
-            ["Je, ni matibabu gani sahihi kwa herpes ya buccal?"],
-        ],inputs=[text_input])
         
+        text_button = gr.Button("Use MultiMed")
+        text_button.click(process_and_query, inputs=[input_language, audio_input, image_input, text_input], outputs=[text_output, audio_output])
+
         clear_button = gr.Button("Clear")
-        clear_button.click(clear, inputs=[], outputs=[input_language, audio_input, image_input, text_input])
+        clear_button.click(clear, inputs=[], outputs=[input_language, audio_input, image_input, text_output, audio_output])
 
     return iface