import subprocess subprocess.run(["python", "-m", "pip", "install", "--upgrade", "pip"]) subprocess.run(["pip", "install", "gradio", "--upgrade"]) 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 # Load model and processor processor = WhisperProcessor.from_pretrained("openai/whisper-large") model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-large") forced_decoder_ids = processor.get_decoder_prompt_ids(language="italian", task="transcribe") # Custom preprocessing function def preprocess_audio(audio_data): # Apply any custom preprocessing to the audio data here if needed # Ensure that the input data is a valid format for the model processed_data = processor(audio_data, return_tensors="pt", padding=True, truncation=True) return processed_data # Function to perform ASR on audio data def transcribe_audio(input_features): # Generate token ids predicted_ids = model.generate(**input_features) # Decode token ids to text transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True) return transcription[0] # Create Gradio interface audio_input = gr.Audio() gr.Interface(fn=transcribe_audio, inputs=audio_input, outputs="text").launch()