Create app.py

app.py

@@ -0,0 +1,63 @@
+import gradio as gr
+import whisper
+from TTS.api import TTS
+import os
+import tempfile
+import numpy as np
+from moviepy.editor import *
+
+def clone_voice(text, reference_audio_path):
+    # Initialize Coqui TTS with a voice cloning model
+    tts = TTS(model_name="tts_models/multilingual/multi-dataset/your_tts", progress_bar=False)
+    
+    # Generate speech with cloned voice
+    output_path = "cloned_voice.wav"
+    tts.tts_to_file(text=text, speaker_wav=reference_audio_path, language="en", file_path=output_path)
+    
+    return output_path
+
+def generate_video_resume(resume_text, photo_path, reference_audio=None):
+    # Clone voice if reference audio is provided
+    if reference_audio:
+        audio_path = clone_voice(resume_text, reference_audio)
+    else:
+        # Fallback to gTTS
+        tts = gTTS(text=resume_text, lang='en')
+        audio_path = "output.mp3"
+        tts.save(audio_path)
+    
+    # Create lip-sync video (simplified example)
+    video_path = "output.mp4"
+    os.system(f"ffmpeg -loop 1 -i {photo_path} -i {audio_path} -c:v libx264 -tune stillimage -c:a aac -b:a 192k -pix_fmt yuv420p -shortest {video_path}")
+    
+    return video_path
+
+with gr.Blocks() as demo:
+    gr.Markdown("# 🎤 Voice My Resume (with Voice Cloning)")
+    
+    with gr.Tab("Standard Voice"):
+        text_input1 = gr.Textbox(label="Paste Resume Text")
+        image_input1 = gr.Image(label="Upload Photo", type="filepath")
+        generate_btn1 = gr.Button("Generate with Default Voice")
+    
+    with gr.Tab("Clone Your Voice"):
+        text_input2 = gr.Textbox(label="Paste Resume Text")
+        image_input2 = gr.Image(label="Upload Photo", type="filepath")
+        audio_input = gr.Audio(label="Upload 10s Reference Audio", type="filepath")
+        generate_btn2 = gr.Button("Generate with Cloned Voice")
+    
+    video_output = gr.Video(label="Result")
+    
+    generate_btn1.click(
+        fn=generate_video_resume,
+        inputs=[text_input1, image_input1],
+        outputs=video_output
+    )
+    
+    generate_btn2.click(
+        fn=generate_video_resume,
+        inputs=[text_input2, image_input2, audio_input],
+        outputs=video_output
+    )
+
+demo.launch()