Speech data packet loss compensation method and device

Abstract

The embodiment of the invention discloses a speech data packet loss compensation method and device which are used for compensating for the lost speech data packet. The method comprises the following steps of: when the lost frame is detected, extracting the pitch period from the received speech data; calculating the backward phase of the received last frame of normal speech data; calculating the forward phase of the lost frame according to the backward phase; and capturing the last frame of data with the same phase as the forward phase from the received data as the compensation data.

Description

technical field [0001] The invention relates to the field of communication technology, in particular to a method and device for compensating voice data packet loss. Background technique [0002] During the transmission of voice data packets in the Internet Protocol (IP, Internet Protocol) network, in order to ensure the real-time performance of data transmission, unreliable transmission methods are generally used for data transmission, for example, the User Datagram Protocol (UDP, User Datagram Protocol) transmission method, there must be packet loss. How to reduce the degradation of voice quality caused by network packet loss is an important research topic for IP networks to carry voice data. [0003] In the prior art, the following two methods are usually adopted: [0004] Existing technology 1. Data packet redundancy technology: at the sending end, multiple copies of the same data packet are sent, as long as one copy of the data is not lost, the original data can be com...

AI Technical Summary

Problems solved by technology

[0006] However, in the above-mentioned prior art, due to the need to send multiple copies of the same data packet, the occupation of the network bandwidth will increase exponentially, which may cause the network to deteriorate further, and the network packet loss may be continuous and bursty. Multiple copies of data may be lost at the same time, so the lost packet cannot be recovered at the receiving end, which will reduce the voice quality and cause delay due to packet loss

In the above-mentioned prior art 2, the compensation data packet is synthesized from the front and rear frames, and the data of the next frame must be received to compensate. If continuous packet loss occurs, the compensation effect will be poor, resulting in a large delay

Images