Vocal fold pathology assessment using AM autocorrelation analysis of the Teager energy operator

Traditional speech processing methods for laryngeal pathology assessment assume linear speech production, with measures derived from an estimated glottal flow waveform. They normally require the speaker to achieve complete glottal closure, which for many vocal fold pathologies cannot be accomplished. To address this, a nonlinear signal processing approach is proposed which employs a differential Teager energy operator and the energy separation algorithm to obtain formant AM and FM modulations from bandpass-filtered speech recordings. A new speech measure is proposed, based on parameterization of the autocorrelation envelope of the AM response. Using a cubic model of the autocorrelation envelope, a 3D space is formed to assess changes in speech quality. This approach is shown to achieve exemplary detection performance for a set of muscular tension dysphonias. Unlike flow characterization using numerical solutions of Navier-Stokes equations, this method is extremely computationally attractive, requiring only NlogN+8N multiplications and N square roots for N samples, and is therefore suitable for real-time applications due to its computational simplicity. The new non-invasive method shows conclusively that a fast, effective digital speech processing technique can be developed for vocal fold pathology assessment, without the need for (i) direct glottal flow, estimation or (ii) complete glottal closure by the speaker. The proposed method also confirms that alternative nonlinear methods can begin to address the limitations of previous linear approaches for speech pathology assessment