Update app.py

app.py

@@ -1,77 +1,24 @@
 import os
 import unicodedata
-from datasets import load_dataset, Audio
 from transformers import pipeline
 import gradio as gr
-import torch
-
-############### HF ###########################
-
-HF_TOKEN = os.getenv("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")
-
-hf_writer = gr.HuggingFaceDatasetSaver(HF_TOKEN, "Urdu-ASR-flags")
-
-############## DagsHub ################################
-
-Model = "kingabzpro/wav2vec2-large-xls-r-300m-Urdu"
-# This is not working because Huggingface has completely changed the git server. 
-# from dagshub.streaming import install_hooks
-# install_hooks()
-
-############## Inference ##############################
 
+# Set the Hugging Face token
+HF_TOKEN = "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"
+os.environ["HF_TOKEN"] = HF_TOKEN
 
+# Define the ASR function
 def asr(audio):
-
-    asr = pipeline("automatic-speech-recognition", model=Model)
+    asr = pipeline("automatic-speech-recognition", model="kingabzpro/wav2vec2-large-xls-r-300m-Urdu")
     prediction = asr(audio, chunk_length_s=30)
-    return unicodedata.normalize("NFC",prediction["text"])
-
-
-################### Gradio Web APP ################################
+    return unicodedata.normalize("NFC", prediction[0]["text"])
 
+# Define the Gradio interface
 title = "Urdu Automatic Speech Recognition"
+description = "This model performs automatic speech recognition for the Urdu language."
+input_audio = gr.inputs.Audio(source="microphone", type="file", label="Record your voice")
+output_text = gr.outputs.Textbox(label="Transcription")
 
-description = """
-<p>
-<center>
-This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice dataset.
-<img src="https://huggingface.co/spaces/kingabzpro/Urdu-ASR-SOTA/resolve/main/Images/cover.jpg" alt="logo" width="550"/>
-</center>
-</p>
-"""
-
-article = "<p style='text-align: center'><a href='https://dagshub.com/kingabzpro/Urdu-ASR-SOTA' target='_blank'>Source Code on DagsHub</a></p><p style='text-align: center'><a href='https://huggingface.co/blog/fine-tune-xlsr-wav2vec2' target='_blank'>Fine-tuning XLS-R for Multi-Lingual ASR with 🤗 Transformers</a></p></center><center><img src='https://visitor-badge.glitch.me/badge?page_id=kingabzpro/Urdu-ASR-SOTA' alt='visitor badge'></center></p>"
-
-examples = [["Sample/sample1.mp3"], ["Sample/sample2.mp3"], ["Sample/sample3.mp3"]]
-
-
-Input = gr.Audio(
-    source="microphone",
-    type="filepath",
-    label="Please Record Your Voice",
-)
-Output = gr.Textbox(label="Urdu Script")
-
-
-def main():
-    iface = gr.Interface(
-        asr,
-        Input,
-        Output,
-        title=title,
-        allow_flagging="manual",
-        flagging_callback=hf_writer,
-        description=description,
-        article=article,
-        examples=examples,
-        theme='JohnSmith9982/small_and_pretty'
-    )
-
-    iface.launch(enable_queue=True)
-
-
-# enable_queue=True,auth=("admin", "pass1234")
+gr.Interface(fn=asr, inputs=input_audio, outputs=output_text, title=title, description=description).launch()
 
-if __name__ == "__main__":
-    main()
+# Ensure to replace "YOUR_HF_TOKEN" with your actual Hugging Face token.