On the Harmonic-plus-Noise decomposition of speech

In this paper, we present an improved method for decomposing the speech signal into the harmonic and the noise parts. The improvements make the method more accurate and robust by including a new pitch detection algorithm. In separating the noise and the harmonic components, this improved harmonic-plus-noise (H+N) decomposition method uses a weighted least-squares method aiming at maximizing the harmonic part. Tests show that the method performs reliably and robustly