Abstract

Recording of a usual mobile phone communication was compared to recordings made in 7 situations in which speech communication differed from the usual one due to noisy ambient, microphone position, etc. Changes of variabilities and average values of long-term formant distributions of 5 vowels of 6 speakers were observed. The first formant of vowel /i/ most often changed variability. The greatest variability had the second formant of vowel /u/. Because of that, speaker recognition based on these two vowels should be carefully analyzed.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call