Create app.py

app.py

@@ -0,0 +1,53 @@
+import os
+import uuid
+import shutil
+import gradio as gr
+from TTS.api import TTS
+
+MODEL_NAME = "tts_models/multilingual/multi-dataset/xtts_v2"
+
+print("🚀 Starting server...")
+print("⬇️ Downloading/loading model only once...")
+
+tts = TTS(
+    model_name=MODEL_NAME,
+    gpu=False  # CPU only for HF Spaces free tier
+)
+
+print("✅ Model loaded. Ready for requests.")
+
+OUTPUT_DIR = "outputs"
+os.makedirs(OUTPUT_DIR, exist_ok=True)
+
+def generate_tts(text, voice_file):
+    if not text or voice_file is None:
+        return None, "❌ Please provide both text and voice sample."
+
+    out_name = f"{uuid.uuid4().hex}.wav"
+    out_path = os.path.join(OUTPUT_DIR, out_name)
+
+    tts.tts_to_file(
+        text=text,
+        speaker_wav=voice_file,
+        language="hi",   # change if needed
+        file_path=out_path
+    )
+
+    return out_path, "✅ Audio generated!"
+
+demo = gr.Interface(
+    fn=generate_tts,
+    inputs=[
+        gr.Textbox(label="Text", placeholder="Enter text here..."),
+        gr.Audio(type="filepath", label="Upload Voice Sample (WAV)")
+    ],
+    outputs=[
+        gr.Audio(label="Generated Audio"),
+        gr.Textbox(label="Status")
+    ],
+    title="XTTS Voice Cloning (CPU, No API Key)",
+    description="Upload a voice sample + text → get cloned voice audio. Model loads once and is reused."
+)
+
+if __name__ == "__main__":
+    demo.launch(server_name="0.0.0.0", server_port=7860)