Pitch estimation using models of voiced speech on three levels

We present an algorithm for estimating the fundamental fre-quency in speech signals. Our approach incorporates mod-els of voiced speech on three levels. First, we estimate the pitch for each time frame based on its harmonic structure us-ing non-negative matrix factorization. The second level uti-lizes temporal pitch continuity to extract partial pitch con-tours. Thirdly, we incorporate statistics of the succession of voiced segments to aggregate partial contours to the final con-tour of an utterance. We evaluate our approach on the Keele database. The experimental results show the robustness of our method for noisy speech, and the good performance for clean speech in comparison with state-of-the-art algorithms. Index Terms — pitch estimation, speech analysis, matrix decomposition 1