This research article presented and focused on recognizing speakers through multi-speaker speeches. The participation of several speakers includes every conference, talk or discussion. This type of talk has different problems as well as stages of processing. Challenges include the unique impurity of the surroundings, the involvement of speakers, speaker distance, microphone equipment etc. In addition to addressing these hurdles in real time, there are also problems in the treatment of the multi-speaker speech. Identifying speech segments, separating the speaking segments, constructing clusters of similar segments and finally recognizing the speaker using these segments are the common sequential operations in the context of multi-speaker speech recognition. All linked phases of speech recognition processes are discussed with relevant methodologies in this article. This entire article will examine the common metrics, methods and conduct. This paper examined the algorithm of speech recognition system at different stages. The voice recognition systems are built through many phases such as voice filter, speaker segmentation, speaker idolization and the recognition of the speaker by 20 speakers.
Read full abstract