Long term stability of tracheoesophageal voice quality