Efficient approach to speaker diarization using voice characteristics extraction - View it on GitHub