This work addresses the detection & characterization of irregular phonation in spontaneous speech. While published work tackles this problem as a two-hypothesis problem only in regions of speech with phonation, this work focuses on distinguishing aperiodicity due to frication from that due to irregular voicing. This work also deals with correction of a current pitch tracking algorithm in regions of irregular phonation, where most pitch trackers fail to perform well. Relying on the detection of regions of irregular phonation, an acoustic parameter is developed in order to characterize these regions for speaker identification applications. The detection performance of the algorithm on a clean speech corpus (TIMIT) is seen to be 91.8%, with the percentage of false detections being 17.42%. On telephone speech corpus (NIST 98) database, the detection performance is 89.2%, with the percentage of false detections being 12.8%. The pitch detection accuracy increased from 95.4% to 98.3% for TIMIT, and from94.8% to 97.4% for NIST 98 databases. The creakiness parameter was added to a set of seven acoustic parameters for speaker identification on the NIST 98 database, and the performance was found to be enhanced by 1.5% for female speakers and 0.4% for male speakers for a population of 250 speakers.