• Ms. Ashwini R. Deokate
• Akanksha Salankar
• Divyani Ninawe
• Mrunali Gatade
• Vishakha Awachat

A revolutionary Fully Supervised Speaker Diarization System is introduced, employing a sophisticated Recurrent Neural Network (RNN) algorithm. Our purpose is to address the limitations of traditional unsupervised methods by leveraging labelled data to achieve simultaneous speaker segmentation and counting. The proposed system showcases its effectiveness through experimental results on a diverse dataset, achieving an impressive speaker counting accuracy of 95%. Unlike conventional approaches, our system offers enhanced adaptability and accuracy in real-world scenarios, presenting a promising solution for applications in speech processing and audio analytic. The system architecture, training procedures, and comprehensive experimental results are outlined, demonstrating the potential of our proposed Fully Supervised Speaker Diarization System to revolutionize the field.

Speaker Diarization, Recurrent Neural Network (RNN), Audio Segmentation, Deep Learning, Speech Processing.

"VoiceParseAI: Speaker Diarization", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.f426-f432, April-2024, Available :http://www.jetir.org/papers/JETIR2404546.pdf

"VoiceParseAI: Speaker Diarization", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. ppf426-f432, April-2024, Available at : http://www.jetir.org/papers/JETIR2404546.pdf