Voice abstract forming method and voice abstract forming system for conference

Abstract

The invention discloses a voice abstract forming method and a voice abstract forming system for a conference, and relates to the field of voice recognition. According to the invention, the speaking position, the identity information, the personnel data and other information of a speaker in a conference are analyzed, and then the weight coefficient of the speaker is determined. As a result, candidate key speaking segments corresponding to different speakers can be obtained by using different preset strategies according to the weight coefficients of speakers. Furthermore, according to the characteristics of the speaking content, for example, the large-probability interval of the important content of the utterance on a speaking time axis, key transitional words and key connection words in theimportant content of statements, and the like, a set of candidate key speaking segments is intercepted. The intercepted set of candidate key speaking segments is processed and then a set of audio / video segments for forming a voice abstract is obtained. As a result, more content is extracted from important statements, while less content is extracted from non-important statements. The content of afinally formed abstract is more reasonable. The more effective help is provided for users.

Description

technical field [0001] The invention relates to a method and a system for forming a speech abstract of a conference, in particular to a method and a system for automatically synthesizing a conference abstract in speech form by identifying key speech contents of different speakers. Background technique [0002] Nowadays, there are countless meetings held every day. In order to record the content of the meeting, it is usually possible to use a video camera or a recording pen to record the video or audio data of the meeting site, save the video data or audio data as a multimedia file, and save it through playback. You can watch or listen to the content of the meeting at any time, or manually convert it into a recorded text afterwards to meet the needs of memorandum and training. In addition, special recorders are usually arranged at the meeting site or the participants themselves take notes by notebook computers or handwriting, etc., to record the content of the meeting held. ...

AI Technical Summary

Problems solved by technology

[0006] However, although Patent Document 1 can identify the speaking participants and associate the key content of the speech with the participants for display, the method of extracting their speech information for all participants is the same, and it cannot be based on different participants. Selectively extract the speech content of different participants in different situations

In an actual meeting, the importance of different participants is usually different. If the same method is used to extract speech information for each participant, it is possible to extract too much information for the speech content of unimportant participants, resulting in a waste of resources. , and too little information is extracted for important participants' speech content, resulting in information omission

Moreover, in Patent Document 1, the content of the statement made by the speaker is identified and transcribed, and after the key points are determined, information such as text is generated and displayed for users to watch and read, which loses the advantages of the voice file itself

In addition, Patent Document 1 judges which participant is speaking by comparing the participant's image with template images of the faces of speakers and non-speakers, and then the computer determines speech participation by applying a hidden Markov model to a summary record of the participant's voice waveform. The voiceprint of the participant, the process of identifying the participant is more complicated and less efficient

[0007] For patent document 2, it is necessary to extract the key information of the venue, combine the key information of the venue into key index points, and combine multiple key index points based on multiple time points into a conference summary. On the other hand, it is easily affected by external environmental factors, its key index points do not necessarily represent the important information of the meeting, and the meeting summary formed by it may not be accurate

Images