Update app.py

app.py

@@ -234,6 +234,8 @@ async def transcript_to_speech(transcript_text, voice, rate, pitch):
                         if current_audio_duration > time_difference:
                             intended_start_time = previous_end_time_ms
                         # print(f"intended_start_time:{intended_start_time}")
+                else:
+                    intended_start_time = start_time
 
                 timed_audio_segments.append({'start': intended_start_time, 'audio': combined_line_audio})
                 previous_end_time_ms = max(previous_end_time_ms, intended_start_time + current_audio_duration)