change to ipfs link

app.py

@@ -11,6 +11,7 @@ import uuid
 from dotenv import load_dotenv
 import tempfile
 import shutil
+from pydub import AudioSegment
 
 # Load environment variables from a .env file if it exists.
 # This is particularly useful for local development.
@@ -54,73 +55,85 @@ def is_url(string):
 
 @app.route('/api/voice-transfer', methods=['POST'])
 def voice_transfer():
-    if 'voice_file' not in request.files:
-        return jsonify({"error": "No voice file part"}), 400
-    
-    file = request.files['voice_file']
-    
-    if file.filename == '':
-        return jsonify({"error": "No selected file"}), 400
-        
     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)
-            
-            text_input = request.form.get('text')
-            app.logger.info(f"Received text input: {text_input}")
-            if not text_input:
-                return jsonify({"error": "No text or text_url provided"}), 400
+        file = None
+        voice_file_path = None
 
-            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:
-                app.logger.info("Input is not a URL, using as raw text.")
-                text_content = text_input
-
-            # --- Perform Voice Cloning using Fish Audio ---
-            app.logger.info("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."
-                )
-            except Exception as e:
-                app.logger.error(f"Error during voice cloning with Fish Audio: {e}")
-                return jsonify({"error": "Failed to generate voice file."}), 500
-            
-            app.logger.info(f"Successfully generated audio file at: {output_file_path}")
-            
-            # 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'
+        # # 1. Check for file upload
+        # if 'voice_file' in request.files:
+        #     file = request.files['voice_file']
+        #     if file.filename == '':
+        #         return jsonify({"error": "No selected file"}), 400
+        #     voice_file_path = os.path.join(temp_dir, file.filename)
+        #     file.save(voice_file_path)
+        # 2. Check for file URL
+        if 'voice_file_url' in request.form:
+            file_url = request.form.get('voice_file_url')
+            if not file_url:
+                return jsonify({"error": "No voice_file_url provided"}), 400
+            response = requests.get(file_url)
+            if response.status_code != 200:
+                return jsonify({"error": "Failed to download voice file from URL"}), 400
+            # Save as webm first
+            webm_path = os.path.join(temp_dir, "input.webm")
+            with open(webm_path, "wb") as f:
+                f.write(response.content)
+            # Convert to mp3
+            mp3_path = os.path.join(temp_dir, "input.mp3")
+            audio = AudioSegment.from_file(webm_path, format="webm")
+            audio.export(mp3_path, format="mp3")
+            voice_file_path = mp3_path
+        else:
+            return jsonify({"error": "No voice file or voice_file_url provided"}), 400
+
+        text_input = request.form.get('text')
+        app.logger.info(f"Received text input: {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:
+            app.logger.info("Input is not a URL, using as raw text.")
+            text_content = text_input
+
+        # --- Perform Voice Cloning using Fish Audio ---
+        app.logger.info("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:
+            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:
+            app.logger.error(f"Error during voice cloning with Fish Audio: {e}")
+            return jsonify({"error": "Failed to generate voice file."}), 500
+
+        app.logger.info(f"Successfully generated audio file at: {output_file_path}")
+
+        with open(output_file_path, 'rb') as f:
+            audio_buffer = io.BytesIO(f.read())
+
+        return send_file(
+            audio_buffer,
+            as_attachment=True,
+            download_name=output_filename,
+            mimetype='audio/mpeg'
+        )
 
     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)