app.py

import subprocess


subprocess.run(["python", "-m", "pip", "install", "--upgrade", "pip"])
subprocess.run(["pip", "install", "gradio", "--upgrade"])
subprocess.run(["pip", "install", "soundfile"])
subprocess.run(["pip", "install", "numpy"])
subprocess.run(["pip", "install", "pydub"])
subprocess.run(["pip", "install", "openai"])

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

# 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

# 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(preprocess=preprocess_audio)
gr.Interface(fn=transcribe_audio, inputs=audio_input, outputs="text").launch()