Can i fine tune "silero_stt" model on my custom dataset?

1

i load model like:

model, decoder, utils = torch.hub.load(repo_or_dir='snakers4/silero-models', model='silero_stt', language='en', device='cuda')

and try to fine tune it with :

for epoch in range(100):
    model.train()
    for batch in data_loader:
        optimizer.zero_grad()
        input, targets = batch
        output = model(input)
        loss = criterion(output, targets)
        loss.backward()
        optimizer.step()
    print(f"Epoch {epoch}: Loss {loss.item()}")