Spectral amplitude warping (SAW) for noise spectrum shaping in audio coding

@article{Lefebvre1997SpectralAW,
  title={Spectral amplitude warping (SAW) for noise spectrum shaping in audio coding},
  author={Roch Lefebvre and Claude Laflamme},
  journal={1997 IEEE International Conference on Acoustics, Speech, and Signal Processing},
  year={1997},
  volume={1},
  pages={335-338 vol.1},
  url={https://api.semanticscholar.org/CorpusID:12255626}
}
A new approach to shape the coding noise in speech and audio coders, called spectral amplitude warping (SAW), consists essentially of a pre- and post-processing which apply a nonlinear transformation to the signal short-term spectrum prior to, and after, encoding.
View on IEEE
mirlab.org
10 Citations
Highly Influential Citations
2
Background Citations
4
Methods Citations
2

Figures and Tables from this paper

10 Citations

Shaping coding noise with frequency-domain companding

This work presents an approach to shape the coding noise in speech and audio coders, which is based on pre-processing and post-processing, called spectral amplitude warping (SAW), which is accomplished by using a constant companding function.

Observations on Frequency-Domain Companding for Audio Coding

    S. Voran
    Computer Science, Engineering
  • 1998
This paper offers additional theoretical background and empirical results on frequency-domain companding, and demonstrates empirically that this technique colors white coding noise so that it is better masked by audio signals, resulting in higher perceived audio quality.

Delay-free lossy audio coding using shelving pre- and post-filters

A delay-free audio coding scheme based on ADPCM with adaptive pre- and post-filtering is presented, which significantly improves the audio quality compared to a plain AD PCM codec, as underlined by objective measurements.

A method and system of pre and post-processing an audio signal for transmission over a channel strongly disturbs

A method and system adapted to modify an audio signal or speech comprising a step where the frequency spectrum S (k) of the signal is transformed by applying a non-lineaire.Le function method…

Waveform quantization of speech using Gaussian mixture models

Waveform quantization of speech using Gaussian mixture models (GMM) is proposed, and high dimensional vector quantizers (VQ) that efficiently exploit the redundancy are constructed based on the GMM parameters.

MPEG audio bit rate scaling on coded data domain

The simulation results show that the re-quantization process provides a very high conversion efficiency and a nearly equal sound quality can be obtained as directly coding from PCM by reflecting the psychoacoustic model.

Coding bit rate scalability of MPEG audio data

Three algorithms for bit rate scaling on coded MPEG data domains are described to cut higher frequency components until the target data rate is satisfied and a requantization process where a quantization step in each subband is modified is used.

Performance of Spectral Amplitude Warp based WDFTC in a Noisy Phoneme and Word Recognition Tasks

The study shows that the PMVDR and WDFTC SAW achieve recognition performance superior to the MFCC and WDFTC in noisy conditions.

A partial decorrelation scheme for improved predictive open loop quantization with noise shaping

The signal to noise ratio of the reconstructed signal can be increased for open loop quantization compared to conventional prediction analysis, and the perceptual encoding quality benefits from a moderate spectral shaping of the quantization noise in the decoded signal.

Compander systems with adaptive preemphasis/deemphasis using linear prediction

A new class of compander systems is proposed that combines conventional broad-band companders with adaptive filtering based on linear prediction. This allows not only for reduction, but also spectral…

5 References

Advances in speech and audio compression

Current activity in speech compression is dominated by research and development of a family of techniques commonly described as code-excited linear prediction (CELP) coding, which offer a quality versus bit rate tradeoff that significantly exceeds most prior compression techniques.

Code-excited linear prediction(CELP): High-quality speech at very low bit rates

A code-excited linear predictive coder in which the optimum innovation sequence is selected from a code book of stored sequences to optimize a given fidelity criterion, indicating that a random code book has a slight speech quality advantage at low bit rates.

Coding of wideband speech

Advances in speech and audio compression : Data compression

This paper presents research and development of a family of techniques commonly described as code-exited linear prediction (CELP) coding, which exploit models of speech production and auditory perception and offer a quality versus bit rate tradeoff that significantly exeeds most prior compression techniques.

7 kHz audio coding within 64 kbit/s

    X. Maître
    Computer Science, Engineering
    IEEE J. Sel. Areas Commun.
  • 1988
The recently approved CCITT Recommendation G.722 on 7-kHz audio coding within 64 kb/s is described, followed by a high-level description of the recommended algorithm and a brief discussion of the detailed arithmetical choices that have been made.