Text this: Robust speaker recognition in noisy environments /