Update app.py

app.py

@@ -1,24 +1,40 @@
-import gradio as gr
+from flask import Flask, request, send_file, jsonify
 from TTS.api import TTS
 import os
+import tempfile
 
+# Accept TOS
 os.environ["COQUI_TOS_AGREED"] = "1"
 
+# Initialize TTS model once
 tts = TTS("tts_models/multilingual/multi-dataset/xtts_v2").to("cpu")
 
-def clone(text, audio):
-    output_path = "./output.wav"
-    tts.tts_to_file(text=text, speaker_wav=audio, language="en", file_path=output_path)
-    return output_path
+app = Flask(__name__)
 
-iface = gr.Interface(
-    fn=clone,
-    inputs=[gr.Textbox(label="Text"), gr.Audio(type="filepath", label="Voice Reference")],
-    outputs=gr.Audio(type="filepath")
-)
+@app.route("/clone", methods=["POST"])
+def clone_route():
+    try:
+        # Get text and audio file from request
+        text = request.form.get("text")
+        audio_file = request.files.get("audio")
 
-# Launch and capture the interface info
-server, _, _ = iface.launch(server_name="0.0.0.0", server_port=7860, share=True)
+        if not text or not audio_file:
+            return jsonify({"error": "Missing text or audio file"}), 400
 
-# Print the public shareable URL
-print(f"Gradio link: {server.local_url} / {server.share_url}")
+        # Save uploaded audio temporarily
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp_audio:
+            audio_path = tmp_audio.name
+            audio_file.save(audio_path)
+
+        # Generate cloned voice
+        output_path = tempfile.NamedTemporaryFile(delete=False, suffix=".wav").name
+        tts.tts_to_file(text=text, speaker_wav=audio_path, language="en", file_path=output_path)
+
+        # Return generated audio
+        return send_file(output_path, mimetype="audio/wav", as_attachment=True, download_name="output.wav")
+
+    except Exception as e:
+        return jsonify({"error": str(e)}), 500
+
+if __name__ == "__main__":
+    app.run(host="0.0.0.0", port=5000, debug=True)