AES standard for audio forensics – Speech Collection Guidelines for Speaker Recognition: Interviewing at a Temporary Location

Project scope: These guidelines are intended to be one of a series and covers only one scenario – the collection of speech samples for speaker recognition at a temporary, non-laboratory location. One example of this would be the collection of speech samples from subjects during some type of field activity. The field activity could be associated with a range of purposes, such as law enforcement, intelligence, military or sociological. These guidelines presume portable resources and somewhat limited time to perform the collection. It also assumes that the interviewer is fluent in the subject’s language, or that an interpreter is present who is both fluent in the interviewer’s and subject’s languages and cognizant of the goals of the interview. These guidelines do not deal with details related to the handling, transmission, storage, or preservation of collected data. Specifically: • It does not deal with any issues related to collection, storage, or protection of personal information. • It does not recommend how to protect speaker recognition analysts from seeing or hearing subject identifying information which could result in biased analysis results. These guidelines also do not deal with possibly important concerns such as personnel and equipment security, power sources for equipment, etc. The alleviation of such highly situation specific concerns is the responsibility of the guideline’s user or their agency.
Assigned to: SC-03-12, Status:
Intent: Review, Initiated: 2022-02-20, Target: 2027
Current goal: Revised Standard, Goal date: Continuing