| 摘要 |
FIELD: physics. ^ SUBSTANCE: essence of the method and device for transmitting speech signals (SS) lies in restoration of the envelope of a speech signal at the receiving side. For this purpose phase keying (PK) by 180 is achieved not through the clipped speech signal, but its time derivative. At the receiving side, alternating-sign pulses of the clipped time derivative of the speech signal are time discretised, resulting in oscillation with delta modulation, which is then discretised through time integration and low-frequency filtration, obtaining the transmitted speech signal with its envelope as a result. With the aim of eliminating non-linear distortions from clipping the differentiation operation, it is not the speech signal (broadband) which is clipped, but a single-band (narrowband) signal formed from the speech signal, which is then linearly transferred once more into the voice band through coherent detection. ^ EFFECT: higher quality of clipping speech signal to maximum possible value after amplitude recovery in the clipped speech signal and elimination of non-linear distortions during clipping. ^ 2 cl, 4 dwg |
