System for concealing missing audio waveforms

摘要

In one embodiment, a method can include: (i) establishing an internet protocol (IP) connection; (ii) forming a buffered version of a plurality of voice frame slices from received audio packets; and (iii) when an erasure is detected, performing a packet loss concealment (PLC) to provide a synthesized speech signal for the erasure, where the PLC can include: (a) identifying first and second pitches from the buffered version of the plurality of voice frame slices; and (b) forming the synthesized speech signal by using the first and second pitches, and more if needed, followed by an overlay-add (OLA).