Update app.py

app.py

@@ -1,7 +1,7 @@
 import os
 import torch
 import gradio as gr
-from TTS.tts.configs.xtts_config import XttsConfig  # ✅ Required for deserialization
+from TTS.tts.configs.xtts_config import XttsConfig  
 from TTS.api import TTS
 
 # ✅ Accept Coqui License Automatically
@@ -10,39 +10,46 @@ os.environ["COQUI_TOS_AGREED"] = "1"
 # ✅ Allow `XttsConfig` in PyTorch's safe globals
 torch.serialization.add_safe_globals([XttsConfig])
 
-# ✅ Force full checkpoint deserialization
-def safe_load_checkpoint(model_path):
-    return torch.load(model_path, map_location="cpu", weights_only=False)  # ✅ Fix PyTorch 2.6 issue
-
-# ✅ Initialize XTTS Model (CPU-only)
-tts = TTS(model_name="tts_models/multilingual/multi-dataset/xtts_v2").to("cpu")
-
-
-
+# ✅ Initialize XTTS Model with Streaming Enabled (Faster on CPU)
+tts = TTS(model_name="tts_models/multilingual/multi-dataset/xtts_v2", stream_inference=True, progress_bar=False).to("cpu")
 
+print("✅ Model loaded successfully!")
+print(f"✅ Supported Languages: {tts.languages.keys()}")  # Debugging step
 
+# ✅ Function for Voice Cloning
 def generate_cloned_voice(text, reference_audio, language):
     output_path = "output.wav"
 
-    # Generate cloned speech with language specification
-    tts.tts_to_file(text=text, speaker_wav=reference_audio, file_path=output_path, language=language)
+    if not text.strip():
+        return "Error: Please enter some text.", None
+
+    if not reference_audio:
+        return "Error: Please upload a reference audio file.", None
 
-    return output_path
+    try:
+        print("🔄 Processing voice cloning...")
+        tts.tts_to_file(text=text, speaker_wav=reference_audio, file_path=output_path, language=language)
+        print("✅ Voice cloning complete!")
 
-print(tts.languages)
+        return output_path
+    except Exception as e:
+        print(f"❌ Error: {str(e)}")
+        return f"Error: {str(e)}", None
 
-# Create the Gradio interface
+# ✅ Create the Gradio Interface
 interface = gr.Interface(
     fn=generate_cloned_voice,
     inputs=[
         gr.Textbox(label="Enter Translated Text"),
         gr.Audio(label="Upload Reference Audio", type="filepath"),
-        gr.Dropdown(["en", "fr", "de", "es", "it"], label="Select Target Language", value="en")
+        gr.Dropdown(list(tts.languages.keys()), label="Select Target Language", value="en")
     ],
     outputs=gr.Audio(label="Generated Cloned Voice"),
     title="Free Voice Cloning API",
-    description="Upload a sample voice and input text. Select a language, and the system will generate the text in the same voice."
+    description="Upload a sample voice and input text. Select a language, and the system will generate the text in the same voice.",
+    allow_flagging="never",
+    concurrency_limit=1  # Prevents multiple processes from running at once
 )
 
-# Launch the Gradio app
+# ✅ Launch the Gradio App
 interface.launch()