Speaker diarisation is the process of separating out speakers in a recording. Our software was developed in response to a real-life requirement to separate out speakers in police interviews and works best with two or three different speakers. It can be used in the pre-processing of recordings to separate out speakers of interest for further analysis, or when building a voice database.
Windows | API
Provides an easy means to separate out the speech of individual speakers from recordings that contain multiple speakers. Quick access for redaction, disguise, or intelligence gathering.