Update app.py

app.py

@@ -11,56 +11,15 @@ from dotenv import load_dotenv
 load_dotenv()
 
 # API Keys
-A_KEY = os.getenv("A_KEY")
 B_KEY = os.getenv("B_KEY")
 
 # URLs
 API_URL = os.getenv("API_URL")
 UPLOAD_URL = os.getenv("UPLOAD_URL")
 
-def get_voices():
-    url = "https://api.elevenlabs.io/v1/voices"
-    headers = {
-        "Accept": "application/json",
-        "xi-api-key": A_KEY
-    }
-    
-    response = requests.get(url, headers=headers)
-    if response.status_code != 200:
-        return []
-    return [(voice['name'], voice['voice_id']) for voice in response.json().get('voices', [])]
-
 def get_video_models():
     return [f for f in os.listdir("models") if f.endswith((".mp4", ".avi", ".mov"))]
 
-def text_to_speech(voice_id, text, session_id):
-    url = f"https://api.elevenlabs.io/v1/text-to-speech/{voice_id}"
-    
-    headers = {
-        "Accept": "audio/mpeg",
-        "Content-Type": "application/json",
-        "xi-api-key": A_KEY
-    }
-    
-    data = {
-        "text": text,
-        "model_id": "eleven_turbo_v2_5",
-        "voice_settings": {
-            "stability": 0.5,
-            "similarity_boost": 0.5
-        }
-    }
-    
-    response = requests.post(url, json=data, headers=headers)
-    if response.status_code != 200:
-        return None
-    
-    # Save temporary audio file with session ID
-    audio_file_path = f'temp_voice_{session_id}.mp3'
-    with open(audio_file_path, 'wb') as audio_file:
-        audio_file.write(response.content)
-    return audio_file_path
-
 def upload_file(file_path):
     with open(file_path, 'rb') as file:
         files = {'fileToUpload': (os.path.basename(file_path), file)}
@@ -139,15 +98,12 @@ def combine_audio_video(video_path, audio_path, output_path):
 
     subprocess.run(cmd, check=True)
 
-def process_video(voice, model, text, progress=gr.Progress()):
+def process_video(model, audio_file, progress=gr.Progress()):
     session_id = str(uuid.uuid4())  # Generate a unique session ID
-    progress(0, desc="Generating speech...")
-    audio_path = text_to_speech(voice, text, session_id)
-    if not audio_path:
-        return None, "Failed to generate speech audio."
     
     progress(0.2, desc="Processing video...")
     video_path = os.path.join("models", model)
+    audio_path = audio_file.name  # Get the path of the uploaded audio file
     
     try:
         progress(0.3, desc="Uploading files...")
@@ -188,36 +144,29 @@ def process_video(voice, model, text, progress=gr.Progress()):
             return output_path, f"Used fallback method. Original error: {str(e)}"
         except Exception as fallback_error:
             return None, f"All methods failed. Error: {str(fallback_error)}"
-    finally:
-        # Cleanup
-        if os.path.exists(audio_path):
-            os.remove(audio_path)
 
 def create_interface():
-    voices = get_voices()
     models = get_video_models()
     
     with gr.Blocks() as app:
-        gr.Markdown("# JSON Train")
+        gr.Markdown("# Video Lip Sync")
         with gr.Row():
             with gr.Column():
-                voice_dropdown = gr.Dropdown(choices=[v[0] for v in voices], label="Select", value=voices[0][0] if voices else None)
-                model_dropdown = gr.Dropdown(choices=models, label="Select", value=models[0] if models else None)
-                text_input = gr.Textbox(label="Enter text", lines=3)
+                model_dropdown = gr.Dropdown(choices=models, label="Select Video Model", value=models[0] if models else None)
+                audio_input = gr.Audio(label="Upload Audio File", type="filepath")
                 generate_btn = gr.Button("Generate Video")
             with gr.Column():
                 video_output = gr.Video(label="Generated Video")
                 status_output = gr.Textbox(label="Status", interactive=False)
         
-        def on_generate(voice_name, model_name, text):
-            voice_id = next((v[1] for v in voices if v[0] == voice_name), None)
-            if not voice_id:
-                return None, "Invalid voice selected."
-            return process_video(voice_id, model_name, text)
+        def on_generate(model_name, audio_file):
+            if not audio_file:
+                return None, "Please upload an audio file."
+            return process_video(model_name, audio_file)
         
         generate_btn.click(
             fn=on_generate,
-            inputs=[voice_dropdown, model_dropdown, text_input],
+            inputs=[model_dropdown, audio_input],
             outputs=[video_output, status_output]
         )
     
@@ -225,4 +174,4 @@ def create_interface():
 
 if __name__ == "__main__":
     app = create_interface()
-    app.launch()
+    app.launch()