How to prevent the output from containing strange characters?

Beginners
1

Hello to everyone,

I’m using t5 to do summarization tasks in english, however, always at the end of the outputs strange symbols are generated.

this is the code:

summary_ids = model.generate(tokenized_text,
                         num_beams=3,
                         no_repeat_ngram_size=2,
                         min_length=300,
                        max_length=600)
 output = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
 print ("\n\nSummarized text: \n",output)

At the end of the output there are symbols like:

á là és ê e †ô óà un­à une­estànèà­ê­m­–as­ô­sè­1­á­2­al­

Is this normal? Is there any way to prevent it?