Update app.py

app.py

@@ -2,16 +2,13 @@ import gradio as gr
 import edge_tts
 import asyncio
 import tempfile
-import os
 from moviepy.editor import AudioFileClip
 import re
 
-# Get all available voices
 async def get_voices():
     voices = await edge_tts.list_voices()
     return {f"{v['ShortName']} - {v['Locale']} ({v['Gender']})": v['ShortName'] for v in voices}
 
-# Text to speech function
 async def text_to_speech(text, voice, rate, pitch):
     if not text.strip():
         return None, gr.Warning("Please enter the text to convert.")
@@ -22,57 +19,51 @@ async def text_to_speech(text, voice, rate, pitch):
     rate_str = f"{rate:+d}%"
     pitch_str = f"{pitch:+d}Hz"
     communicate = edge_tts.Communicate(text, voice_short_name, rate=rate_str, pitch=pitch_str)
+    
     with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmp_file:
         tmp_path = tmp_file.name
         await communicate.save(tmp_path)
+    
     return tmp_path, None
 
-# Generate SRT based on estimated timing
-def generate_srt(text, speech_rate, max_words_per_line):
-    # Clean up input text
-    text = re.sub(r'\s+', ' ', text.strip())  # Remove excessive whitespace
-
-    # Split into words
+def generate_srt(text, audio_duration, max_words_per_line):
+    # Eliminate extra spaces and split into words
+    text = re.sub(r'\s+', ' ', text.strip())
     words = text.split()
-    
-    # Calculate timing for each line
     srt_lines = []
     current_line = []
-    current_time = 0.0  # Start time in seconds
     total_words = len(words)
-
+    
     for i, word in enumerate(words):
         current_line.append(word)
-        
-        # Calculate current line length
+        # Create a line if we reach the max words per line or at the end of the text
         if len(current_line) >= max_words_per_line or i == total_words - 1:
-            # Create SRT entry
             line_text = ' '.join(current_line)
-            duration = len(line_text.split()) / speech_rate  # Estimate duration based on speech rate
+            # Adjust duration proportionally
+            duration = audio_duration * (len(current_line) / total_words)
+            start_time = (sum(len(' '.join(srt_lines[j].split()[2:])) for j in range(len(srt_lines))) / total_words) * audio_duration if srt_lines else 0
             
-            # Format timing
-            start_time = current_time
-            end_time = current_time + duration
+            end_time = start_time + duration
             
+            # Formatting time for SRT
             start_time_str = f"{int(start_time // 3600):02}:{int((start_time % 3600) // 60):02}:{int(start_time % 60):02},{int((start_time % 1) * 1000):03}"
             end_time_str = f"{int(end_time // 3600):02}:{int((end_time % 3600) // 60):02}:{int(end_time % 60):02},{int((end_time % 1) * 1000):03}"
             
             srt_lines.append(f"{len(srt_lines) + 1}\n{start_time_str} --> {end_time_str}\n{line_text}\n")
-            
-            # Move to the next line
             current_line = []
-            current_time += duration  # Update current time
-
+    
     return ''.join(srt_lines)
 
-# Gradio interface function
-def tts_interface(text, voice, rate, pitch, speech_rate, max_words_per_line):
+def tts_interface(text, voice, rate, pitch, max_words_per_line):
     audio_path, warning = asyncio.run(text_to_speech(text, voice, rate, pitch))
     if warning:
         return None, None, warning
     
+    # Calculate audio duration
+    audio_duration = AudioFileClip(audio_path).duration  # Get duration in seconds
+    
     # Generate SRT file
-    srt_content = generate_srt(text, speech_rate, max_words_per_line)
+    srt_content = generate_srt(text, audio_duration, max_words_per_line)
     srt_path = audio_path.replace('.mp3', '_subtitle.srt')
     
     with open(srt_path, 'w') as f:
@@ -80,7 +71,6 @@ def tts_interface(text, voice, rate, pitch, speech_rate, max_words_per_line):
 
     return audio_path, srt_path, None
 
-# Create Gradio app
 async def create_demo():
     voices = await get_voices()
     
@@ -91,7 +81,6 @@ async def create_demo():
             gr.Dropdown(choices=[""] + list(voices.keys()), label="Select Voice", value=""),
             gr.Slider(minimum=-50, maximum=50, value=0, label="Rate Adjustment (%)", step=1),
             gr.Slider(minimum=-20, maximum=20, value=0, label="Pitch Adjustment (Hz)", step=1),
-            gr.Slider(minimum=100, maximum=300, value=150, label="Speech Rate (words per minute)", step=1),
             gr.Slider(minimum=3, maximum=8, value=5, label="Max Words per Line", step=1),
         ],
         outputs=[
@@ -107,7 +96,6 @@ async def create_demo():
     
     return demo
 
-# Run the app
 if __name__ == "__main__":
     demo = asyncio.run(create_demo())
     demo.launch()