fix directory issue

app.py

@@ -9,6 +9,8 @@ import io
 from fish_audio import clone_voice_with_fish
 import uuid
 from dotenv import load_dotenv
+import tempfile
+import shutil
 
 # Load environment variables from a .env file if it exists.
 # This is particularly useful for local development.
@@ -60,56 +62,65 @@ def voice_transfer():
     if file.filename == '':
         return jsonify({"error": "No selected file"}), 400
         
-    if file:
-        # Save the uploaded file temporarily
-        # In a production environment, you'd want a more robust storage solution
-        upload_folder = 'uploads'
-        if not os.path.exists(upload_folder):
-            os.makedirs(upload_folder)
+    temp_dir = tempfile.mkdtemp()
+    try:
+        if file:
+            # Save the uploaded file in the temporary directory
+            voice_file_path = os.path.join(temp_dir, file.filename)
+            file.save(voice_file_path)
             
-        voice_file_path = os.path.join(upload_folder, file.filename)
-        file.save(voice_file_path)
-        
-        text_input = request.form.get('text')
-        print(text_input)
-        if not text_input:
-            return jsonify({"error": "No text or text_url provided"}), 400
+            text_input = request.form.get('text')
+            print(text_input)
+            if not text_input:
+                return jsonify({"error": "No text or text_url provided"}), 400
 
-        text_content = ""
-        if is_url(text_input):
-            text_content = get_text_from_url(text_input)
-            if text_content is None:
-                return jsonify({"error": "Failed to retrieve or parse content from URL"}), 400
-        else:
-            print("Not URL")
-            text_content = text_input
+            text_content = ""
+            if is_url(text_input):
+                text_content = get_text_from_url(text_input)
+                if text_content is None:
+                    return jsonify({"error": "Failed to retrieve or parse content from URL"}), 400
+            else:
+                print("Not URL")
+                text_content = text_input
 
-        # --- Perform Voice Cloning using Fish Audio ---
-        print("Starting voice cloning process with Fish Audio...")
-        
-        output_dir = 'outputs_v2'
-        os.makedirs(output_dir, exist_ok=True)
-        output_filename = f"output_cloned_{uuid.uuid4().hex}.mp3"
-        output_file_path = os.path.join(output_dir, output_filename)
+            # --- Perform Voice Cloning using Fish Audio ---
+            print("Starting voice cloning process with Fish Audio...")
+            
+            output_filename = f"output_cloned_{uuid.uuid4().hex}.mp3"
+            output_file_path = os.path.join(temp_dir, output_filename)
 
-        try:
-            # Note: For best results, provide an accurate transcript of the reference audio.
-            # Since we don't get it from the user, we can pass a generic placeholder or an empty string.
-            clone_voice_with_fish(
-                text=text_content,
-                reference_audio_path=voice_file_path,
-                output_path=output_file_path,
-                reference_text="This is a reference audio for voice cloning."
+            try:
+                # Note: For best results, provide an accurate transcript of the reference audio.
+                # Since we don't get it from the user, we can pass a generic placeholder or an empty string.
+                clone_voice_with_fish(
+                    text=text_content,
+                    reference_audio_path=voice_file_path,
+                    output_path=output_file_path,
+                    reference_text="This is a reference audio for voice cloning."
+                )
+            except Exception as e:
+                print(f"Error during voice cloning with Fish Audio: {e}")
+                return jsonify({"error": "Failed to generate voice file."}), 500
+            
+            # Read the generated file into a memory buffer
+            with open(output_file_path, 'rb') as f:
+                audio_buffer = io.BytesIO(f.read())
+            
+            # Send the buffer as a file
+            return send_file(
+                audio_buffer,
+                as_attachment=True,
+                download_name=output_filename,
+                mimetype='audio/mpeg'
             )
-        except Exception as e:
-            print(f"Error during voice cloning with Fish Audio: {e}")
-            return jsonify({"error": "Failed to generate voice file."}), 500
-        
-        try:
-            return send_file(output_file_path, as_attachment=True)
-        except Exception as e:
-            app.logger.error(f"Error sending file: {e}")
-            return jsonify({"error": "Failed to send audio file."}), 500 
+
+    except Exception as e:
+        app.logger.error(f"An unexpected error occurred: {e}")
+        return jsonify({"error": "An internal server error occurred."}), 500
+    finally:
+        # Clean up the temporary directory and its contents
+        if temp_dir and os.path.exists(temp_dir):
+            shutil.rmtree(temp_dir)
 
 # if __name__ == '__main__':
 #     app.run(debug=True, port=5001)