Update app.py

app.py

@@ -1,32 +1,24 @@
-import subprocess
-
-subprocess.run(["pip", "install", "datasets"])
-subprocess.run(["pip", "install", "transformers"])
-subprocess.run(["pip", "install", "torch", "torchvision", "torchaudio", "-f", "https://download.pytorch.org/whl/torch_stable.html"])
-
 import gradio as gr
-from transformers import WhisperProcessor, WhisperForConditionalGeneration
+import whisper
 
-# Load model and processor
-processor = WhisperProcessor.from_pretrained("openai/whisper-large")
-model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-large")
-model.config.forced_decoder_ids = None
 
-# Custom preprocessing function
-def preprocess_audio(audio_data):
-    # Apply any custom preprocessing to the audio data here if needed
-    return processor(audio_data, return_tensors="pt").input_features
+def transcribe_audio(audio_file):
+    model = whisper.load_model("base")
+    result = model.transcribe(audio_file)
+    return result["text"]
 
-# Function to perform ASR on audio data
-def transcribe_audio(input_features):
-    # Generate token ids
-    predicted_ids = model.generate(input_features)
+
+def main():
+    audio_input = gr.inputs.Audio(source="upload", type="filepath")
+    output_text = gr.outputs.Textbox()
     
-    # Decode token ids to text
-    transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
+    iface = gr.Interface(fn=transcribe_audio, inputs=audio_input, 
+                         outputs=output_text, title="Audio Transcription App",
+                         description="Upload an audio file and hit the 'Submit'\
+                             button")
     
-    return transcription[0]
+    iface.launch()
+
 
-# Create Gradio interface
-audio_input = gr.Audio(preprocess=preprocess_audio)
-gr.Interface(fn=transcribe_audio, inputs=audio_input, outputs="text").launch()
+if __name__ == '__main__':
+    main()