Method of a voice source acoustic analysis in real time

The problem of non-invasive research of the speech apparatus vocal function by the announcer's speech signal is considered. A new method of acoustic analysis of a pulse-type voice source based on a two-stage measurement procedure has been developed. The first stage of measurements provides for filtering the voice excitation signal of the vocal tract, and the second stage – converting this signal into a final pulse sequence synchronous with the main tone of the speech signal. An example of technical (software) implementation of the developed method is considered, estimates of its computational complexity and speed are given. The ability of the method to be used in the soft (with a delay of hundredths of a second) real time mode has been established. A full-scale experiment has been set up and conducted using the author's software. It is shown that at limited intervals of vocalization of the speech signal the developed method guarantees stability of the repetition rate and shape of excitation impulses, which is valuable from the point of view of the accuracy of measurements of all the main parameters of the speech vocal source: from the fundamental frequency to the amplitude disturbances (flickering) of the source pulses. The obtained results will be useful in developing new and upgrading existing algorithms and technologies for speech signal synthesis and digital speech transmission over low-speed communication channels, as well as medical diagnostics and voice therapy systems.