Update app.py

app.py

@@ -1,42 +1,48 @@
 import os
-import gradio as gr
 import tempfile
+import gradio as gr
 import librosa
 import soundfile as sf
 
-# ========== Trust Required Configs for PyTorch 2.6+ ==========
+# ===== Step 1: Allowlist Required Classes for PyTorch >= 2.6 =====
 from torch.serialization import add_safe_globals
 from TTS.tts.configs.xtts_config import XttsConfig
-from TTS.tts.models.xtts import XttsAudioConfig
+from TTS.tts.models.xtts import XttsAudioConfig, XttsArgs
 from TTS.config.shared_configs import BaseDatasetConfig
 
-add_safe_globals([XttsConfig, XttsAudioConfig, BaseDatasetConfig])
+add_safe_globals([
+    XttsConfig,
+    XttsAudioConfig,
+    XttsArgs,
+    BaseDatasetConfig
+])
 
-# ========== Agree to Coqui TTS Terms ==========
+# ===== Step 2: Agree to Coqui TTS Terms of Service =====
 os.environ["COQUI_TOS_AGREED"] = "1"
 
-# ========== Initialize the Coqui TTS Model ==========
+# ===== Step 3: Load the Coqui XTTS Model =====
 from TTS.api import TTS
+
 tts = TTS(
     model_name="tts_models/multilingual/multi-dataset/xtts_v2",
     progress_bar=True,
-    gpu=False  # Set to True if using a CUDA-compatible GPU
+    gpu=False  # Set to True if using CUDA
 )
 
-# ========== Inference Function ==========
+# ===== Step 4: Define Voice Cloning Inference Function =====
 def text_to_speech_clone(text, voice_sample):
     if voice_sample is None:
         return "Please provide a voice sample audio.", None
 
-    # Load voice sample
+    # Load the voice sample audio file
     sample_wav, sample_rate = librosa.load(voice_sample, sr=22050)
 
-    # Save sample to temp file
+    # Save sample temporarily in correct format
     with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as tmp_sample:
         sf.write(tmp_sample.name, sample_wav, sample_rate)
         voice_sample_path = tmp_sample.name
 
-    # Generate cloned Hindi speech
+    # Generate cloned Hindi speech and save it to a temp file
     with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as tmp_output:
         tts.tts_to_file(
             text=text,
@@ -48,7 +54,7 @@ def text_to_speech_clone(text, voice_sample):
 
     return output_path
 
-# ========== Gradio Interface ==========
+# ===== Step 5: Gradio UI Interface =====
 iface = gr.Interface(
     fn=text_to_speech_clone,
     inputs=[
@@ -58,10 +64,10 @@ iface = gr.Interface(
     outputs=gr.Audio(type="filepath", label="Generated Cloned Speech"),
     title="Hindi Text-to-Speech with Voice Cloning",
     description=(
-        "Generate Hindi speech from text with voice cloning capability.\n"
-        "Provide a short Hindi voice sample (5–10 seconds) to clone its voice tone."
+        "यह ऐप हिंदी टेक्स्ट से वॉयस क्लोनिंग के साथ स्पीच जेनरेट करता है।\n"
+        "एक छोटी सी हिंदी आवाज़ की रिकॉर्डिंग (5-10 सेकंड) अपलोड करें, और यह उसी आवाज़ में टेक्स्ट पढ़कर सुनाएगा।"
     )
 )
 
-# ========== Launch ==========
+# ===== Step 6: Launch the Web App =====
 iface.launch()