Digitally Disguised Voices
One of the newer challenges in audio forensics is the investigation of voice recordings that are suspect of being manipulated by utilizing digital signal processing. This paper presents a number of voice manipulating tools available. Several tools are developed for musical purposes. However, they are eventually applied to talkers’ voices as well. This implies both post-processing real time processing. The voice alterations are demonstrated and the influence on acoustic parameters normally examined in connection with forensic voice analysis and authentication are presented. This includes the effect of pitch shifting, pitch modeling, formant relocation, removal of voiced sounds, and its’ influence on background sounds (ENF).
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is temporarily free for AES members.