Update app.py

app.py

@@ -209,7 +209,7 @@ def generate_audio_kokoro(text, lang, selected_voice):
     global kokoro_tts  # Access the preloaded model
     if os.path.exists(f"audio_{lang}.wav"):
         os.remove(f"audio_{lang}.wav")  
-    yield None
+    
     lang_code = SUPPORTED_TTS_LANGUAGES.get(lang, "a")  # Default to English
     #generator = kokoro_tts(text, voice="bm_george", speed=1, split_pattern=r'\n+')
     generator = kokoro_tts(text, voice=selected_voice, speed=1, split_pattern=r'\n+')
@@ -220,7 +220,7 @@ def generate_audio_kokoro(text, lang, selected_voice):
     
     output_file = f"audio_{lang}.wav"
     sf.write(output_file, full_audio, 24000)  # Save as WAV file
-    yield output_file
+    return output_file
 	
 ### 5️⃣ Chunk-Based Summarization
 def split_text_with_optimized_overlap(text, max_tokens=1024, overlap_tokens=25):
@@ -341,14 +341,18 @@ with gr.Blocks() as demo:
     #process_summary_button.click(hierarchical_summarization, inputs=[extracted_text], outputs=[summary_output])
 	
     process_audio_button.click(
-        lambda text, summary, lang, voice, tts_choice: generate_audio_kokoro(
-            summary if tts_choice == "Summary Audio" else text, lang, voice
+        lambda text, summary, lang, voice, tts_choice: (
+            None,  # Clear previous audio
+            generate_audio_kokoro(
+                summary if tts_choice == "Summary Audio" else text, lang, voice
+            )
         ),
         inputs=[extracted_text, summary_output, detected_lang, voice_selection, tts_option],
-        outputs=[full_audio_output],
+        outputs=[full_audio_output, full_audio_output],  # Clear first, then display new audio
         show_progress=True
     )
 
+
     process_ner_button.click(
         #extract_entities_with_stanza, 
 	    extract_entities_with_gliner,