ACELP at 5.3 kbit/s & MP-MLQ at 6.3 kbit/s
The coder is based on the principles of linear prediction analysis-by synthesis coding and attempts to minimise a perceptually weighted error signal. It operates on a digital input signal specified with 16 bit (PCM linear) at 8 kHz sampling rate. The decoder converts the received signal after decoding back to an analogue signal. The coder can be used for compressing the speech or other audio signals (not represented faithfully) component of multimedia services at a very low bit rate, two-bit rates are associated with 5.3 and 6.3 kbit/s, as part of the overall H.234 family of standards. The higher bit rate offers greater quality, whereby the lower offers good quality and provides system designers with additional flexibility. It is possible to switch between the two rates at any frame boundary. Variable rate operation using discontinuous transmission and noise fill during non-speech intervals is possible. The excitation signal for the high rate coder is Multi-Pulse Maximum Likelihood Quantization (MP-MLQ) and for the low rate coder is Algebraic-Code-Excited Linear Prediction (ACELP). The coder was optimise for high quality at low rates using limited amount of complexity. The frame size is 30 ms and additional look ahead of 7.5 ms is required. Therefore the total algorithmic delay of 37.5 ms results. Additional delay is caused due to processing, transmission and buffering of the multiplexing protocol.