Audio Forensics
Background
The analysis of audio data in relation to meta-information, which can be extracted from the acoustic signals in addition to spoken language, is an area that is still mostly unexplored. In addition to the already well-researched transcription or identification of speakers, the environments in which the audio data was recorded could be investigated.
Aim
Besides the reconstruction of the recording location (e.g., regional delimitation of the location, indoor or outdoor, size, etc.), it would also be possible to analyze any information that can be generated via the recording device, e.g., a smartphone microphone. Depending on the uniqueness, information could also be extracted for precise identification, recognition, or a restriction of the range of speakers (noise generators).
Disruptive Risk Research
It is currently questionable whether it is even possible to assign the recording location based on audio data of varying quality. It is not possible to estimate whether the proposed methods will be applicable in the context of criminal investigations. The legally secure provability of potential research results represents a significant challenge. From a scientific and technical perspective, the project must be considered high-risk research.
Contact
Always up to date? Subscribe to our scientific newsletter and you will quickly find out which calls for proposals we have just published as well as other interesting information from the field of research and innovation.