• Pragati V. Thawani
• Mohit Khatri
• Dr. Nitya S.

Speaker recognition is a vital task in biometric authentication and speech analysis systems. Traditional methods depend heavily on handcrafted features and are often limited by performance in text-independent scenarios. In this paper, we propose a deep learning-based approach utilizing multi-layer Convolutional Neural Networks (CNNs) to perform text- independent speaker recognition. Our model learns hierarchical feature representations directly from Mel-spectrograms, allowing robust performance across varying speech inputs. We evaluate our system on the VoxCeleb1 dataset and demonstrate superior accuracy compared to traditional machine learning and baseline deep learning models.

Speaker recognition, convolutional neural networks, deep learning, biometric authentication, text-independent

"Text-Independent Speaker Recognition Using Multi- Layer Convolutional Neural Networks", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 10, page no.c593-c598, October-2025, Available :http://www.jetir.org/papers/JETIR2510276.pdf

"Text-Independent Speaker Recognition Using Multi- Layer Convolutional Neural Networks", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 10, page no. ppc593-c598, October-2025, Available at : http://www.jetir.org/papers/JETIR2510276.pdf