This document summarizes a research paper on using i-vectors for artist recognition in music. The proposed method extracts i-vectors from songs to obtain a compact representation that captures artist variability. It uses a Gaussian mixture model (GMM) to calculate statistics from frame-level features, then performs factor analysis to extract i-vectors from the GMM supervectors. Experiments on a dataset of 20 artists show the method achieves better artist recognition performance than baselines, and works well across different backends like discriminant analysis and probabilistic linear discriminant analysis.