A.NSI SCTE 24-23-2017 pdf free download – BV32 Speech Codec Specification for Voice over IP Applications in Cable Telephony.
In Figure 1 above, the outer layer (including the two short-term predictors and the short-term noise feedback filter) follows the structure of the conventional NFC codec. The TSNFC structure in Figure 1 is obtained by replacing the simple scalar quantizer in the conventional (single-stage) NFC structure by a “predictive quantizer” that employs long-term prediction and long-term noise spectral shaping. This “predictive quantizer” is represented by the inner feedback loop in Figure 1, including the long-term predictor and long-term noise feedback filter. This inner feedback loop uses an alternative but equivalent conventional NFC structure, where ) (z N l represents the filter whose frequency response is the desired noise shape for long-term noise spectral shaping. In the outer layer, the short-term noise feedback filter ) (z F s is usually chosen as a bandwidth-expanded version of the short-term predictor ) (z P s . The choice of different NFC structures in the outer and inner layers is based on complexity consideration. By combining two stages of NFC in a nested loop, the TSNFC in Figure 1 can reap the benefits of both short-term and long-term prediction and also achieve short-term and long-term noise spectral shaping at the same time.
It is natural and straightforward to use a scalar quantizer in Figure 1. However, to achieve better coding efficiency, a vector quantizer is used in BV32. In the Vector Quantization (VQ) codebook search, the ) (n u vector cannot be generated before the VQ codebook search starts. Due to the feedback structure in Figure 1, the elements of ) (n u from the second element on will depend on the vector-quantized version of earlier elements. Therefore, the VQ codebook search is performed by trying out each of the candidate codevectors in the VQ codebook (i.e. fixing a candidate ) (n uq vector first), calculating the corresponding ) (n u vector and the corresponding VQ error ) ( ) ( ) ( n uq n u n q. The VQ codevector that minimizes the energy of ) (n q within the current vector time span is chosen as the winning codevector, and the corresponding codebook index becomes part of the encoder output bit stream for the current speech frame.
The TSNFC decoder structure is simply a quantizer decoder followed by the two feedback filter structures involving the long-term predictor and the short-term predictor, respectively, shown on the right half of Figure 1. Thus, the TSNFC decoder is similar to the decoders of other predictive coding techniques such as Adaptive Predictive Coding (APC), Multi-Pulse Linear Predictive Coding (MPLPC), and Code-Excited Linear Prediction (CELP).
2.2 Overview of the BV32 Codec
The BV32 codec is a purely forward-adaptive TSNFC codec. It operates at an input sampling rate of 16 kHz and an encoding bit rate of 32 kb/s, or 2 bits per sample. The BV32 uses a frame size of 5 ms, or 80 samples. There is no look ahead. Therefore, the total algorithmic buffering delay is just the frame size itself, or 5 ms. The main design goal of BV32 is to make the coding delay and the codec complexity as low as possible, while maintaining essentially transparent output speech quality.A.NSI SCTE 24-23 pdf download.