Text this: Speaker recognition based on adaptive boosting and parallel multiple expert models /