Inserting Fractions with Multiple Denominators using Speech to Text