Update app.py

app.py

@@ -1,26 +1,22 @@
 import os
+from datasets import load_dataset
 from TTS.api import TTS
-from datasets import load_dataset, Audio
 import gradio as gr
 
-# Automatically accept Coqui license
+# Accept Coqui license automatically
 os.environ["COQUI_TOS_AGREED"] = "1"
 
-# Load dataset from Hugging Face
+# Load your dataset and retrieve the voice sample path
 dataset = load_dataset("Emmylahot12/nnamdi", split="train")
-dataset = dataset.cast_column("audio", Audio())
-
-# Validate audio exists
-if not dataset or not dataset[0]["audio"] or not dataset[0]["audio"]["path"]:
+if dataset[0]["audio"] is None:
     raise ValueError("Dataset is empty or audio is missing")
 
-# Reference voice sample path
 voice_sample_path = dataset[0]["audio"]["path"]
 
-# Initialize TTS with XTTS v2 model (CPU mode)
+# Initialize the TTS engine (CPU)
 tts = TTS(model_name="tts_models/multilingual/multi-dataset/xtts_v2", gpu=False)
 
-# TTS synthesis function
+# Inference function
 def synthesize(text, language="en"):
     output_path = "output.wav"
     tts.tts_to_file(
@@ -32,16 +28,12 @@ def synthesize(text, language="en"):
     return output_path
 
 # Gradio UI
-interface = gr.Interface(
+gr.Interface(
     fn=synthesize,
     inputs=[
-        gr.Textbox(label="Enter Text"),
-        gr.Dropdown(choices=["en", "fr", "es", "de"], label="Language")
+        gr.Textbox(label="Enter text to synthesize"),
+        gr.Dropdown(["en", "fr", "es"], label="Language", value="en")
     ],
-    outputs=gr.Audio(label="Generated Speech"),
-    title="Nnamdi TTS - Powered by Coqui XTTS v2",
-    description="Custom voice from dataset: Emmylahot12/nnamdi"
-)
-
-if __name__ == "__main__":
-    interface.launch()
+    outputs=gr.Audio(label="Generated Audio"),
+    title="Nnamdi TTS App (XTTSv2)"
+).launch()