Update app.py

output path set

app.py

@@ -44,28 +44,50 @@ async def text_to_speech(text, voice, rate, pitch, output_path):
 # Generate SRT file with specified lines of subtitles
 def generate_srt(words, audio_duration, srt_path, num_lines):
     with open(srt_path, 'w', encoding='utf-8') as srt_file:
-        divisor = len(words) // (5 * num_lines)
-        if divisor == 0:
-            segment_duration = audio_duration  # Use full duration as fallback
-        else:
-            segment_duration = audio_duration / divisor  # Calculate duration per segment
-        
+        segment_duration = audio_duration / (len(words) // (5 * num_lines))  # Average duration for each segment
         current_time = 0
+        
         for i in range(0, len(words), 5 * num_lines):
             lines = []
             for j in range(num_lines):
-                line = ' '.join(words[i + j * 5:i + (j + 1) * 5])
+                line = ' '.join(words[i + j * 5:i + (j + 1) * 5])  # 5 words per line
                 if line:
                     lines.append(line)
-            
+                    
             start_time = current_time
             end_time = start_time + segment_duration
+            
             start_time_str = format_srt_time(start_time)
             end_time_str = format_srt_time(end_time)
             srt_file.write(f"{i // (5 * num_lines) + 1}\n{start_time_str} --> {end_time_str}\n" + "\n".join(lines) + "\n\n")
+            
             current_time += segment_duration
 
     return srt_path
+# def generate_srt(words, audio_duration, srt_path, num_lines):
+#     with open(srt_path, 'w', encoding='utf-8') as srt_file:
+#         divisor = len(words) // (5 * num_lines)
+#         if divisor == 0:
+#             segment_duration = audio_duration  # Use full duration as fallback
+#         else:
+#             segment_duration = audio_duration / divisor  # Calculate duration per segment
+        
+#         current_time = 0
+#         for i in range(0, len(words), 5 * num_lines):
+#             lines = []
+#             for j in range(num_lines):
+#                 line = ' '.join(words[i + j * 5:i + (j + 1) * 5])
+#                 if line:
+#                     lines.append(line)
+            
+#             start_time = current_time
+#             end_time = start_time + segment_duration
+#             start_time_str = format_srt_time(start_time)
+#             end_time_str = format_srt_time(end_time)
+#             srt_file.write(f"{i // (5 * num_lines) + 1}\n{start_time_str} --> {end_time_str}\n" + "\n".join(lines) + "\n\n")
+#             current_time += segment_duration
+
+#     return srt_path
 
 
 def format_srt_time(seconds):
@@ -78,8 +100,9 @@ def format_srt_time(seconds):
     return f"{hours:02}:{minutes:02}:{seconds:02},{millis:03}"
 
 # Text to audio and SRT functionality
-async def text_to_audio_and_srt(text, voice, rate, pitch, num_lines):
-    audio_path, warning = await text_to_speech(text, voice, rate, pitch)
+
+async def text_to_audio_and_srt(text, voice, rate, pitch, num_lines, output_audio_path, output_srt_path):
+    audio_path, warning = await text_to_speech(text, voice, rate, pitch, output_audio_path)
     if warning:
         return None, None, warning
 
@@ -87,17 +110,33 @@ async def text_to_audio_and_srt(text, voice, rate, pitch, num_lines):
     audio_duration = audio_clip.duration
     
     # Generate SRT file based on the entire text
-    base_name = os.path.splitext(audio_path)[0]
-    srt_path = f"{base_name}_subtitle.srt"
     words = text.split()
-    generate_srt(words, audio_duration, srt_path, num_lines)
+    generate_srt(words, audio_duration, output_srt_path, num_lines)
+
+    return audio_path, output_srt_path, None
+# async def text_to_audio_and_srt(text, voice, rate, pitch, num_lines):
+#     audio_path, warning = await text_to_speech(text, voice, rate, pitch)
+#     if warning:
+#         return None, None, warning
+
+#     audio_clip = AudioFileClip(audio_path)
+#     audio_duration = audio_clip.duration
+    
+#     # Generate SRT file based on the entire text
+#     base_name = os.path.splitext(audio_path)[0]
+#     srt_path = f"{base_name}_subtitle.srt"
+#     words = text.split()
+#     generate_srt(words, audio_duration, srt_path, num_lines)
 
-    return audio_path, srt_path, None
+#     return audio_path, srt_path, None
 
 # Gradio interface function
-def tts_interface(text, voice, rate, pitch, num_lines):
-    audio_path, srt_path, warning = asyncio.run(text_to_audio_and_srt(text, voice, rate, pitch, num_lines))
+def tts_interface(text, voice, rate, pitch, num_lines, output_audio_path="output_audio.mp3", output_srt_path="output_subtitle.srt"):
+    audio_path, srt_path, warning = asyncio.run(text_to_audio_and_srt(text, voice, rate, pitch, num_lines, output_audio_path, output_srt_path))
     return audio_path, srt_path, warning
+# def tts_interface(text, voice, rate, pitch, num_lines):
+#     audio_path, srt_path, warning = asyncio.run(text_to_audio_and_srt(text, voice, rate, pitch, num_lines))
+#     return audio_path, srt_path, warning
 
 # Create Gradio app
 async def create_demo():