Previous research on speaker identification has demonstrated the effectiveness of using syllables containing a nasal consonant. In this study, we investigated the contributions of nasal consonants (/m/ and /n/) and vowels (/i/, /e/, /a/, /o/, and /Ɯ/) to identification accuracy by using them separately in speaker identification experiments. Japanese monosyllables with nasal onsets were recorded from 50 male speakers using a condenser microphone. Two recording sessions were held and thus non-contemporaneous speech data were obtained. Nasal consonants and the following vowels were excerpted from the recorded monosyllables, and 30th-order cepstral coefficients were calculated for each as acoustic features. The results revealed that the accuracy of identification using nasal consonants was not as high as that using vowels; more than six nasal tokens for a given speaker needed to be registered in order to match the score afforded by one vowel token for the same speaker. The higher vowels, /Ɯ/, /e/ and /i/, yielded significantly better identification rates than the lower vowels, /a/ and /o/, and the alveolar nasal /n/ was better than the bilabial /m/. We also conducted a factor analysis in order to clarify the effects of the attributes of speakers and speech samples on the differentiation between speakers with similar speech characteristics. Analysis was performed on frequently confused speaker pairs using 11 parameters for vowels and six parameters for nasals. The parameters were selected from various attributes related to the physiological properties of the speakers and the acoustic properties of their speech. The results showed that irregularity in phonation and the degree of vowel nasalisation were among the most influential factors. The physical size of the speakers and the average fundamental frequencies also affected the accuracy of speaker identification.