Create app.py

app.py

@@ -0,0 +1,30 @@
+import gradio as gr
+import onnxruntime as ort
+import numpy as np
+import soundfile as sf
+
+# Load ONNX model
+session = ort.InferenceSession("assets/model.onnx", providers=["CPUExecutionProvider"])
+
+def tts(text):
+    # Encode text to IDs (dummy simple encoding: byte values)
+    input_ids = np.array([list(text.encode("utf-8"))], dtype=np.int64)
+
+    # Run ONNX model
+    audio = session.run(None, {"text": input_ids})[0][0]
+
+    # Normalize audio if needed
+    audio = audio.astype(np.float32)
+    return (44100, audio)
+
+def infer(text):
+    return tts(text)
+
+iface = gr.Interface(
+    fn=infer,
+    inputs=gr.Textbox(label="Input text"),
+    outputs=gr.Audio(label="Output audio"),
+    title="Supertonic TTS (CPU mode)"
+)
+
+iface.launch()