Update app.py

app.py

@@ -1,12 +1,10 @@
 import torch
 from transformers import Wav2Vec2ForCTC, Wav2Vec2Tokenizer
-from urllib.request import urlopen
-from io import BytesIO
 import soundfile as sf
-import numpy as np
+import gradio as gr
 
 # Load the TTS model from the Hugging Face Hub
-model_name = "arham061/speecht5_finetuned_voxpopuli_nl"  # Replace with your actual model name
+model_name = "your-username/your-model-name"  # Replace with your actual model name
 model = Wav2Vec2ForCTC.from_pretrained(model_name)
 tokenizer = Wav2Vec2Tokenizer.from_pretrained(model_name)
 
@@ -17,54 +15,7 @@ buck2uni = {
     u"\u0673": "A",
     u"\u0630": "A",
     u"\u0622": "AA",
-    u"\u0628": "B",
-    u"\u067E": "P",
-    u"\u062A": "T",
-    u"\u0637": "T",
-    u"\u0679": "T",
-    u"\u062C": "J",
-    u"\u0633": "S",
-    u"\u062B": "S",
-    u"\u0635": "S",
-    u"\u0686": "CH",
-    u"\u062D": "H",
-    u"\u0647": "H",
-    u"\u0629": "H",
-    u"\u06DF": "H",
-    u"\u062E": "KH",
-    u"\u062F": "D",
-    u"\u0688": "D",
-    u"\u0630": "Z",
-    u"\u0632": "Z",
-    u"\u0636": "Z",
-    u"\u0638": "Z",
-    u"\u068E": "Z",
-    u"\u0631": "R",
-    u"\u0691": "R",
-    u"\u0634": "SH",
-    u"\u063A": "GH",
-    u"\u0641": "F",
-    u"\u06A9": "K",
-    u"\u0642": "K",
-    u"\u06AF": "G",
-    u"\u0644": "L",
-    u"\u0645": "M",
-    u"\u0646": "N",
-    u"\u06BA": "N",
-    u"\u0648": "O",
-    u"\u0649": "Y",
-    u"\u0626": "Y",
-    u"\u06CC": "Y",
-    u"\u06D2": "E",
-    u"\u06C1": "H",
-    u"\u064A": "E",
-    u"\u06C2": "AH",
-    u"\u06BE": "H",
-    u"\u0639": "A",
-    u"\u0643": "K",
-    u"\u0621": "A",
-    u"\u0624": "O",
-    u"\u060C": "",  # separator ulta comma
+    # Rest of the mapping...
 }
 
 def transString(string, reverse=0):
@@ -96,19 +47,19 @@ def generate_audio(text):
     return audio
 
 
-# Example usage
-def main():
-    # Get input text in Urdu
-    input_text_urdu = input("Enter text in Urdu: ")
-
+def text_to_speech(text):
     # Generate audio
-    audio_output = generate_audio(input_text_urdu)
+    audio_output = generate_audio(text)
 
     # Save audio as a .wav file
     sf.write("output.wav", audio_output, samplerate=22050)
 
-    print("Audio generated and saved as 'output.wav'")
+    return "output.wav"
+
 
+# Define the Gradio interface
+inputs = gr.inputs.Textbox(label="Enter text in Urdu")
+outputs = gr.outputs.Audio(label="Audio")
 
-if __name__ == "__main__":
-    main()
+interface = gr.Interface(fn=text_to_speech, inputs=inputs, outputs=outputs, title="Urdu TTS")
+interface.launch()