A Speaker Segmentation Method Based on Event Detection

Abstract

The invention relates to an event detection based speaker segmentation method which belongs to the technical field of audio signal processing. The method comprises three steps, referred to as, speaker segmentation, event detection and correction of speaker segmentation results. In the speaker segmentation step, the changing points of a speaker in an inputted video are determined based on a currently available distance measuring method. According to the changing points, the video is divided into a plurality of video segments in different lengths, or initialized speaker segmentation results. In the event detection step, based on the time characteristics of events, the event points in the inputted video are detected. In the correction of speaker segmentation results, corrections based on probability statistics are made to the initialized speaker segmentation results so as to obtain the final speaker segmentation results. Compared to the prior art and aimed to intercommunications, the method provided by the invention fully utilizes the information in a video to conduct speaker segmentation so that under the conditions of changed channels, big noise and large numbers of speakers, an enhanced speaker segmentation method becomes available.

Description

technical field [0001] This patent relates to a speaker segmentation method, in particular to a speaker segmentation method based on event detection, which is suitable for speaker segmentation of radio station speech, and belongs to the technical field of audio signal processing. Background technique [0002] With the rapid increase of audio acquisition channels and quantity, audio management has become more and more complex. In recent years, speaker segmentation clustering has gradually become a hot research issue in the world. Many foreign universities and research institutions have carried out related research work. In the speaker recognition evaluation task organized by the National Institute of Standards and Technology (NIST) in 1999, the segmentation and clustering project of two-person telephone speech was added. In 2002, the rich information transcription (Rich Transcription, RT) evaluation officially launched the research on speaker segmentation clustering. [0003...

AI Technical Summary

Problems solved by technology

From the above analysis, it can be seen that the speaker segmentation method does not make full use of other auxiliary information of the audio, and the audio performance is relatively poor for audio with a relatively large noise intensity and a constantly changing channel.

Images