Robust Speaker Recognition In The Presence Of Speech Coding Distortion