Update app.py

app.py

@@ -4,14 +4,14 @@ import asyncio
 import tempfile
 import os
 from moviepy.editor import AudioFileClip
-import speech_recognition as sr
+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 functionality
+# 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.")
@@ -27,120 +27,84 @@ async def text_to_speech(text, voice, rate, pitch):
         await communicate.save(tmp_path)
     return tmp_path, None
 
-# Function to analyze audio and get speech timing
-def analyze_audio(audio_path):
-    recognizer = sr.Recognizer()
-    with sr.AudioFile(audio_path) as source:
-        audio_data = recognizer.record(source)
-        # Recognize speech using Google Web Speech API
-        try:
-            text = recognizer.recognize_google(audio_data)
-            return text
-        except sr.UnknownValueError:
-            return ""
-        except sr.RequestError:
-            return ""
-
-# Generate SRT file based on user preferences
-def generate_srt(words, audio_duration, srt_path, words_per_line, lines_per_paragraph):
-    total_segments = (len(words) // words_per_line) // lines_per_paragraph + 1
-    segment_duration = audio_duration / total_segments  # Calculate duration for each segment
-
-    current_time = 0
-    with open(srt_path, 'w', encoding='utf-8') as srt_file:
-        for i in range(0, len(words), words_per_line):
-            # Gather lines based on the defined words per line
-            lines = words[i:i + words_per_line]
-            line_text = ' '.join(lines)
+# 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
+    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
+        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
+            
+            # Format timing
             start_time = current_time
-            end_time = min(start_time + segment_duration, audio_duration)  # Ensure it doesn't exceed audio duration
+            end_time = current_time + duration
             
-            start_time_str = format_srt_time(start_time)
-            end_time_str = format_srt_time(end_time)
-            srt_file.write(f"{(i // words_per_line) + 1}\n{start_time_str} --> {end_time_str}\n")
-            srt_file.write(f"{line_text}\n\n")
-
-            current_time += segment_duration  # Update current time for the next segment
-
-    return srt_path
-
-def format_srt_time(seconds):
-    millis = int((seconds - int(seconds)) * 1000)
-    seconds = int(seconds)
-    minutes = seconds // 60
-    hours = seconds // 3600
-    minutes %= 60
-    seconds %= 60
-    return f"{hours:02}:{minutes:02}:{seconds:02},{millis:03}"
+            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
 
-# Text to audio and SRT functionality
-async def text_to_audio_and_srt(text, voice, rate, pitch, words_per_line, lines_per_paragraph):
-    # Clean up input text: remove extra spaces and newlines
-    cleaned_text = ' '.join(text.split())
+    return ''.join(srt_lines)
 
-    audio_path, warning = await text_to_speech(cleaned_text, voice, rate, pitch)
+# Gradio interface function
+def tts_interface(text, voice, rate, pitch, speech_rate, max_words_per_line):
+    audio_path, warning = asyncio.run(text_to_speech(text, voice, rate, pitch))
     if warning:
         return None, None, warning
-
-    audio_clip = AudioFileClip(audio_path)
-    audio_duration = audio_clip.duration
     
-    # Analyze audio to get the actual spoken text
-    spoken_text = analyze_audio(audio_path)
-
-    # Generate SRT file based on the entire text
-    base_name = os.path.splitext(audio_path)[0]
-    srt_path = f"{base_name}_subtitle.srt"
-    
-    # Split input text into words
-    words = cleaned_text.split()
+    # Generate SRT file
+    srt_content = generate_srt(text, speech_rate, max_words_per_line)
+    srt_path = audio_path.replace('.mp3', '_subtitle.srt')
     
-    generate_srt(words, audio_duration, srt_path, words_per_line, lines_per_paragraph)
+    with open(srt_path, 'w') as f:
+        f.write(srt_content)
 
     return audio_path, srt_path, None
 
-# Gradio interface function
-def tts_interface(text, voice, rate, pitch, words_per_line, lines_per_paragraph):
-    audio_path, srt_path, warning = asyncio.run(text_to_audio_and_srt(text, voice, rate, pitch, words_per_line, lines_per_paragraph))
-    return audio_path, srt_path, warning
-
 # Create Gradio app
 async def create_demo():
     voices = await get_voices()
     
-    with gr.Blocks() as demo:
-        gr.Markdown(
-            """
-            <h1 style="text-align: center; color: #333;">Text to Speech with Subtitles</h1>
-            <p style="text-align: center; color: #555;">Convert your text to natural-sounding speech and generate subtitles (SRT) for your audio.</p>
-            """, 
-            elem_id="header"
-        )
-
-        with gr.Row():
-            with gr.Column():
-                text_input = gr.Textbox(label="Input Text", lines=5, placeholder="Enter text here...")
-                voice_dropdown = gr.Dropdown(choices=[""] + list(voices.keys()), label="Select Voice", value="")
-                rate_slider = gr.Slider(minimum=-50, maximum=50, value=0, label="Rate Adjustment (%)", step=1)
-                pitch_slider = gr.Slider(minimum=-20, maximum=20, value=0, label="Pitch Adjustment (Hz)", step=1)
-
-                words_per_line = gr.Slider(minimum=3, maximum=8, value=5, label="Words per Line", step=1)
-                lines_per_paragraph = gr.Slider(minimum=1, maximum=5, value=2, label="Lines per Paragraph", step=1)
-                
-                generate_button = gr.Button("Generate Audio and Subtitles", variant="primary")
-
-            with gr.Column():
-                output_audio = gr.Audio(label="Generated Audio", type="filepath")
-                output_srt = gr.File(label="Generated SRT", file_count="single")
-                warning_msg = gr.Markdown(label="Warning", visible=False)
-
-        generate_button.click(
-            fn=tts_interface,
-            inputs=[text_input, voice_dropdown, rate_slider, pitch_slider, words_per_line, lines_per_paragraph],
-            outputs=[output_audio, output_srt, warning_msg]
-        )
-
+    demo = gr.Interface(
+        fn=tts_interface,
+        inputs=[
+            gr.Textbox(label="Input Text", lines=5),
+            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=[
+            gr.Audio(label="Generated Audio", type="filepath"),
+            gr.File(label="Generated Subtitle (.srt)"),
+            gr.Markdown(label="Warning", visible=False)
+        ],
+        title="Edge TTS Text to Speech with SRT",
+        description="Convert text to speech and generate synchronized subtitles based on speech rate.",
+        analytics_enabled=False,
+        allow_flagging=False,
+    )
+    
     return demo
 
 # Run the app