Packet loss is one of the most essential problems in speech communication. It will cause the information loss and uncomfortable for listeners in voice over IP. This investment proposed an approach based on waveform similarity measure using overlap-and-Add algorithm. The waveform similarity overlap-and-add (WSOLA) technique is an effective algorithm to deal with packet loss concealment (PLC). For real-time time communication, the WSOLA algorithm is widely used to deal with the length adaptation and packet loss concealment of speech signal. Time scale modification of audio signal is one of the most essential research topics in data communication, especially in voice of IP (VoIP). Herein, we proposed the dual-side WSOLA that is derived by standard WSOLA. Instead of only exploitation one direction speech data, the proposed method will reconstruct the lost voice data according to the preceding and cascading voice. The dual-side WSOLA can use both the past and future speech signal waveform to reconstruction voice waveform of lost packet. The evaluations show that the quality of the reconstructed speech signal of the dual-side WSOLA is higher than that of the standard WSOLA and GWSOLA on different packet loss rate and length using the metrics: PESQ and MOS. The significant improvement is obtained by dual side information in the proposed method. The proposed dual-side waveform similarity overlap-and-add (DSWSOLA) outperforms the traditional approaches.
Packet Loss Recovery in Broadcast for Real-Time Applications in Dense Wireless Networks
