Add subtitle generation option and guidance for SRT file creation in text_to_speech

app.py

@@ -223,7 +223,8 @@ async def text_to_speech(text, voice, rate, pitch, generate_subtitles=False, upl
     else:
         # Use the existing approach for regular text
         communicate = edge_tts.Communicate(text, voice_short_name, rate=rate_str, pitch=pitch_str)
-        
+        if not generate_subtitles:
+            await communicate.save(audio_path)
         if generate_subtitles:
             with tempfile.NamedTemporaryFile(delete=False, suffix=".srt") as srt_file:
                 subtitle_path = srt_file.name
@@ -336,7 +337,6 @@ async def create_demo():
             with gr.Column(scale=3):
                 text_input = gr.Textbox(label="Input Text", lines=5, value="Hello, how are you doing!")
                 file_input = gr.File(label="Or upload a TXT/SRT file", file_types=[".txt", ".srt"])
-            
             with gr.Column(scale=2):
                 voice_dropdown = gr.Dropdown(
                     choices=[""] + list(voices.keys()),
@@ -354,6 +354,25 @@ async def create_demo():
                     minimum=-20, maximum=20, value=0, label="Pitch Adjustment (Hz)", step=1
                 )
                 subtitle_checkbox = gr.Checkbox(label="Generate Subtitles (.srt)", value=False)
+                gr.Markdown("""
+                    **📝 Subtitle Timing Tip:**
+                    
+                    When creating SRT files for continuous speech, avoid exact matching timestamps between segments.
+                    
+                    **For smoother speech flow:**
+                    ```
+                    1
+                    00:00:00,112 --> 00:00:01,647
+                    Hello how are you doing
+                    
+                    2
+                    00:00:01,617 --> 00:00:02,000
+                    I'm fine
+                    ```
+                    
+                    ✅ Create a small overlap (20-30ms) between segments to prevent pauses
+                    ❌ Avoid exact matching timestamps (where end time = next start time) except you want a pause
+                """)
         
         submit_btn = gr.Button("Convert to Speech", variant="primary")
         warning_md = gr.Markdown(visible=False)