At low bit rates speech is coded frame by frame, each frame of a size of 20-30ms. Perceptually, each frame may be different, depends upon the speech signal properties (voiced or unvoiced). The frames with lower perception may be one of the potential candidates to be dropped and reconstructed to maximize the bandwidth of the systems. This chapter ex

Speech communication is often realised by two ways monologue and dialogue modes. The temporal (time) attributes, such as active and non-active parts of both of these varies because of the nature of the modes. In monologue modes of speech communication activity of each user remains 80% and 20% are the silences and in dialogue (conversational) modes

A low bit rate STDM multiplexer architecture design for thing route channels is provided in this chapter. A thin route communication channel is able to transmit 64Kbps to 128Kbps. Allowing compressed speech using STDM maximize the bandwidth by allowing more users than the capacity ofthe channel. The design provided in this chapter is based upon, co

In packetized speech communication the packet loss occurs. The packets can also be forced to drop in situations when traffic load increase at the multiplexing end. The dropped packets can be substituted at the demultiplexing end. To do so frame substitution techniques can be utilised to mitigate the effects oflost packets. In this chapter packet lo

The design of multiplexer operation can be based upon certain simulation parameters. In this chapter a design of STDM is based upon some parameters considering a 64 Kbps satellite channel. Allowing each user a capacity of 6.4kbps, 10 users can be multiplexed in a normal situation. Using VAD and packet discarding for each of the users allmv more use

Considering low bit rate sources the quality of speech suffers mainly by a factor, such as quantization noise. Multiplexing efficiency is therefore, based upon the encoding sources bit rates. Lower the coding rates, better the efficiency but on the expense of quality. The quality in multiplexing is assessed by two means, objectively and subjectivel

The cyclic frame discarding is an approach in which at least for each user a packet is discarded before attempting to discard a second packet fonn the same. The idea is to avoid consecutive packet discarding from a single user when there is higher load. To do so, a slightly better perfonnance, in tenns of speech quality is achieved than a random pa

A great number ofspeech coding has been developed in recent years and many others are in their advanced level of development. Most of these encoding algorithms use hybrid methodology of wavefonn coding and voice coding commonly knov.m as (vocoding). The speech quality ofthese coders is comparable with Pulse Code Modulation (PCM); a popular encoding

Speech activity and non-activity is a scenario in which each activity (talkspurt) and non-activity (silence) may vary in size. Some talkspurts may be many frames and others may be only a few frames long, considering each frame of speech of 20-30ms. In this chapter a dynamic talker activity analysis is provided through statistical measures to descri

The Digital Speech interpolation (DSI) advantage is signal processing technique, using this technique one can maximise the systems efficiency, such as a telephonic system capability to accommodate more users due to the non-active parts within speech signal. By utilising non activity parts along with Lost Frame Reconstruction (LFR), higher DSI advan