GGSM and VOIP telephonic voice recordings are frequently used in forensic voice comparisons. However, speech compression, enhancement and error correction algorithms are likely to modify speaker specific voice characteristics. We performed high-quality voice recordings and concomitant GSM and VOIP telephonic voice recordings so that clean high-quality and telephonic versions of the same speech are available. Sustained vowel tasks are also included allowing F0 (AVG and STD), Jitter, Shimmer and HNR extraction for pairs of contemporaneous recordings. Twenty female and seventeen male volunteer subjects participated in the study. We describe the voice recording experiments, present results for five vowels and five acoustic parameters for each subject in GSM and VOIP communication, and we discuss the implications of the main results.
Authors:
Fernandes, Vania; Ferreira, Anibal
Affiliation:
University of Porto, Porto, Portugal
AES Conference:
2017 AES International Conference on Audio Forensics (June 2017)
Paper Number:
2-3
Publication Date:
June 6, 2017
Subject:
Speaker Recognition
Click to purchase paper as a non-member or you can login as an AES member to see more options.
No AES members have commented on this paper yet.
To be notified of new comments on this paper you can subscribe to this RSS feed. Forum users should login to see additional options.
If you are not yet an AES member and have something important to say about this paper then we urge you to join the AES today and make your voice heard. You can join online today by clicking here.