Identification of simultaneous speakers

Although the problem of automatic speaker identification has received considerable attention, no work has been made to deal with overlapping speech that involves multiple persons speaking simultaneously. We propose two approaches to automatically identify both single speaker and two speakers in audio recordings that contain overlapping speech. The first approach consists of an overlapping-speech detection component that determines if an audio recording contains overlapping speech, followed by either a single-speaker identifier or a two-speaker identifier. The second approach runs the single-speaker identifier and two-speaker identifier in parallel. Our experiment results demonstrate the feasibility of the two approaches.