Scalable wideband speech coding for IP networks
Koji Seto, Tokunbo Ogunfunmi · 2012
Voice communication over IP networks is inherently less reliable and thus high robustness to packet loss is one of the key features of speech coder for voice over IP (VoIP) applications. This paper presents a scalable wideband speech coder designed especially for IP networks to achieve high packet-loss robustness. The coder adopts a split-band structure, where the input signal sampled at 16 kHz is split into two subbands. Both the lower- and higher-band signals are encoded by the scalable multi-rate internet Low Bit-rate Codec (iLBC). The iLBC inherently possesses high robustness to packet loss and therefore is considered to be more suited for a core-layer coding scheme of scalable wideband coders on VoIP networks. Based on the objective evaluation, the proposed codec provides high robustness to packet loss and achieves slightly higher voice quality than G.729.1 at the bit rate higher than about 24 kbps under clean channel condition.