An overview of the requirements of digital covert audio acquisition (DCAA) is provided and a scalable hybrid coder tuned for the coding of intelligible speech in a covert environment is proposed. The codec incorporates linear predictive coding (LPC) and an M-band discrete wavelet transform (DWT) to offer effective intelligible speech coding in the presence of multiple signal sources at bit rates between 8 kbps and 32 kbps. The results of informal intelligibility testing and an analysis of the algorithm's complexity are presented to demonstrate the performance of the proposed coder.

Hybrid coding of mixed signals for digital covert audio surveillance

Two nonuniform aspects of the line spectrum pair (LSP) linear predictive coding (LPC) parameters are investigated, including nonuniform statistical distributions and spectral sensitivities of adjacent LSP frequency differences. Based upon these two nonuniform properties, a globally optimal scalar quantizer is designed for each differential LSP frequency. The design algorithm is dynamic programming based and minimization of a nontrivial data dependent spectral distortion is adopted as the optimality criterion. At 32 bits/frame, the new LSP quantizer achieves a 1-dB average log spectral distortion, a commonly accepted level for reproducing perceptually transparent spectral information. The quantization performance has also been shown to be robust across different speakers and databases. >

Optimal quantization of LSP parameters

We present a novel way to design biorthogonal and paraunitary linear phase (LPPUFB) filter banks. The square error of the perfect reconstruction condition is expressed in a quadratic form of filter coefficients and the cost function is minimized by solving the linear equation iteratively without nonlinear optimization. With some modifications, the method can be extended to the design of paraunitary filter banks. Using this method, we can design LPPUFB with many channels easily and quickly. Design examples are given to validate the proposed method.

Time-domain design of linear-phase PR filter banks

The aim of combined coding is to compress a mixed signal consisting of speech and music to the highest possible compression ratio and yet maintain user acceptability. This paper proposes a hybrid coder which is capable of coding combined speech and audio signals sampled over the range 16 kHz to 32 kHz. The coder provides high quality audio coding at competitive bit rates over this range of signal sampling frequencies and offers further coding gains when used with speech signals alone.

http://ieeexplore.ieee.org/iel3/5225/14128/00648530.pdf

Hybrid coding of mixed signals for digital covert audio surveillance

References

Optimal quantization of LSP parameters

Time-domain design of linear-phase PR filter banks

Combined coding of audio and speech signals using LPC and the discrete wavelet transform

Related Papers (5)

A two stage hybrid embedded speech/audio coding structure

A multimode transform predictive coder (MTPC) for speech and audio

Arithmetic coding of speech and audio spectra using tcx based on linear predictive spectral envelopes

Packet-loss robust scalable speech coding using the discrete wavelet transform

Hybrid audio coding for speech and audio below medium bit rate