Update app.py

app.py

@@ -47,11 +47,11 @@ def split_text_into_segments(text):
     return segments
 
 # Function to generate SRT with accurate timing per batch and cross-check timing
-async def generate_accurate_srt(batch_text, batch_num, start_offset):
+async def generate_accurate_srt(batch_text, batch_num, start_offset, pitch):
     audio_file = f"batch_{batch_num}_audio.wav"
     
-    # Generate the audio using edge-tts
-    tts = edge_tts.Communicate(batch_text, "en-US-AndrewNeural", rate="-25%")
+    # Generate the audio using edge-tts with pitch adjustment
+    tts = edge_tts.Communicate(batch_text, "en-US-AndrewNeural", rate="-25%", pitch=f"{pitch}Hz")
     await tts.save(audio_file)
 
     # Get the actual length of the audio file
@@ -81,7 +81,7 @@ async def generate_accurate_srt(batch_text, batch_num, start_offset):
     return srt_content, audio_file, start_time  # Return updated start time for cumulative tracking
 
 # Batch processing function with cumulative timing, progress indicator, and final SRT validation
-async def batch_process_srt_and_audio(script_text, progress=gr.Progress()):
+async def batch_process_srt_and_audio(script_text, pitch, progress=gr.Progress()):
     batches = [script_text[i:i+500] for i in range(0, len(script_text), 500)]
     all_srt_content = ""
     combined_audio = AudioSegment.empty()
@@ -89,7 +89,7 @@ async def batch_process_srt_and_audio(script_text, progress=gr.Progress()):
 
     # Process each batch sequentially to ensure proper timing and cumulative offset tracking
     for batch_num, batch_text in enumerate(batches):
-        srt_content, audio_file, end_offset = await generate_accurate_srt(batch_text, batch_num, start_offset)
+        srt_content, audio_file, end_offset = await generate_accurate_srt(batch_text, batch_num, start_offset, pitch)
         all_srt_content += srt_content
 
         # Append the audio of each batch to the combined audio
@@ -129,20 +129,23 @@ async def batch_process_srt_and_audio(script_text, progress=gr.Progress()):
     return final_srt_path, final_audio_path
 
 # Gradio interface function
-async def process_script(script_text):
-    srt_path, audio_path = await batch_process_srt_and_audio(script_text)
+async def process_script(script_text, pitch):
+    srt_path, audio_path = await batch_process_srt_and_audio(script_text, pitch)
     return srt_path, audio_path, audio_path
 
-# Gradio interface setup
+# Gradio interface setup with pitch adjustment slider
 app = gr.Interface(
     fn=process_script,
-    inputs=gr.Textbox(label="Enter Script Text", lines=10),
+    inputs=[
+        gr.Textbox(label="Enter Script Text", lines=10),
+        gr.Slider(label="Pitch Adjustment (Hz)", minimum=-100, maximum=100, step=1, value=0)
+    ],
     outputs=[
         gr.File(label="Download SRT File"),
         gr.File(label="Download Audio File"),
         gr.Audio(label="Play Audio")
     ],
-    description="Upload your script text, and the app will generate audio with en-US-AndrewNeural voice (Rate: -25%) and an accurate SRT file for download."
+    description="HIVEcorp TTS Generator with en-US-AndrewNeural voice (Rate: -25%) and an accurate SRT file for download."
 )
 
 app.launch()