update app.py

app.py

@@ -5,42 +5,56 @@ import torch
 import gradio as gr
 from TTS.api import TTS
 
+# Patch torch.load for compatibility
 old_torch_load = torch.load
 def patched_torch_load(*args, **kwargs):
     kwargs["weights_only"] = False
     return old_torch_load(*args, **kwargs)
 torch.load = patched_torch_load
 
+# Accept Coqui TOS
 os.environ["COQUI_TOS_AGREED"] = "1"
 
+# Model name
 MODEL = "tts_models/multilingual/multi-dataset/xtts_v2"
-
 print("Loading model:", MODEL)
 
+# Detect if GPU available (Hugging Face ZeroGPU = CPU only)
 use_gpu = torch.cuda.is_available()
 tts = TTS(MODEL, gpu=use_gpu)
 
+# Supported languages
 LANGS = [
     "en", "es", "fr", "de", "it", "pt", "pl", "tr", "ru", "nl",
     "cs", "ar", "zh-cn", "ja", "ko", "hu", "hi"
 ]
 
+# Default speaker if no wav provided
 DEFAULT_SPEAKER = tts.speakers[0] if tts.speakers else None
 print("Default speaker:", DEFAULT_SPEAKER)
 
-_ = tts.tts("Hello world, the model is ready!", language="en", speaker=DEFAULT_SPEAKER)
-
 def generate_audio(text, language, speaker_wav):
     if not text or len(text.strip()) < 2:
         return None
 
     out_path = tempfile.mktemp(suffix=".wav")
-    speaker_path = speaker_wav if speaker_wav else None
 
-    if speaker_path:
-        tts.tts_to_file(text=text, speaker_wav=speaker_path, language=language, file_path=out_path)
+    if speaker_wav:
+        # Use reference WAV for voice cloning
+        tts.tts_to_file(
+            text=text,
+            speaker_wav=speaker_wav,
+            language=language,
+            file_path=out_path
+        )
     else:
-        tts.tts_to_file(text=text, speaker=DEFAULT_SPEAKER, language=language, file_path=out_path)
+        # Use built-in default voice
+        tts.tts_to_file(
+            text=text,
+            speaker=DEFAULT_SPEAKER,
+            language=language,
+            file_path=out_path
+        )
 
     return out_path
 
@@ -52,7 +66,7 @@ demo = gr.Interface(
         gr.Audio(label="Speaker reference (optional, WAV)", type="filepath")
     ],
     outputs=gr.Audio(type="filepath", label="Generated speech"),
-    title="XTTS-v2 (multilingual TTS)",
+    title="XTTS-v2 (Multilingual + Voice Cloning)",
     allow_flagging="never",
 )