Upload 2 files

app.py

@@ -0,0 +1,45 @@
+
+import gradio as gr
+from TTS.api import TTS
+import tempfile
+
+# Load XTTS model
+tts = TTS(model_name="tts_models/multilingual/multi-dataset/xtts_v2")
+
+def clone_voice(text, speaker_wav):
+    if speaker_wav is None:
+        return None, "Please upload a reference audio file."
+
+    # Save uploaded audio
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp:
+        tmp.write(speaker_wav.read())
+        speaker_path = tmp.name
+
+    output_path = "cloned_output.wav"
+
+    # Generate audio using XTTS
+    tts.tts_to_file(
+        text=text,
+        speaker_wav=speaker_path,
+        language="zh",
+        file_path=output_path
+    )
+
+    return output_path, "Voice cloning completed successfully."
+
+# Gradio UI
+demo = gr.Interface(
+    fn=clone_voice,
+    inputs=[
+        gr.Textbox(label="Enter Chinese Text"),
+        gr.Audio(label="Upload Speaker Audio (.wav)", type="file")
+    ],
+    outputs=[
+        gr.Audio(label="Cloned Output Audio"),
+        gr.Textbox(label="Status")
+    ],
+    title="XTTS Voice Cloning Demo",
+    description="Upload reference audio and enter Chinese text to generate speech in cloned voice."
+)
+
+demo.launch()

requirements.txt

@@ -0,0 +1,2 @@
+TTS
+gradio