Create app.py

app.py

@@ -0,0 +1,36 @@
+import gradio as gr
+from transformers import pipeline
+import torch
+
+# Load the Whisper model pipeline for speech recognition with optimizations
+model_name = "Vira21/Whisper-Small-Khmer"
+whisper_pipeline = pipeline(
+    "automatic-speech-recognition", 
+    model=model_name,
+    device=0 if torch.cuda.is_available() else -1  # Use GPU if available, otherwise use CPU
+)
+
+def transcribe_audio(audio):
+    try:
+        # Process and transcribe the audio
+        result = whisper_pipeline(audio)["text"]
+        return result
+    except Exception as e:
+        # Handle errors and return an error message
+        return f"An error occurred during transcription: {str(e)}"
+
+# Gradio Interface with optimizations
+interface = gr.Interface(
+    fn=transcribe_audio,
+    inputs=gr.Audio(type="filepath"),  
+    outputs="text",
+    title="Whisper Base Khmer Speech-to-Text",
+    description="Upload an audio file or record your voice to get the transcription in Khmer.",
+    examples=[["Example Audio/126.wav"]],
+    allow_flagging="never"  # Disables flagging to save resources
+)
+
+# Launch the app with queue enabled for better handling on free CPU
+if __name__ == "__main__":
+    interface.queue()  # Enable asynchronous queuing for better performance
+    interface.launch()