Cepstrum

About: Cepstrum is a research topic. Over the lifetime, 3346 publications have been published within this topic receiving 55742 citations.

Perceptually Weighted Mel-Cepstrum Analysis of Speech Based on Psychoacoustic Model

Abstract: This letter proposes a novel approach for mel-cepstral analysis based on the psychoacoustic model of MPEG. A perceptual weighting function is developed by applying cubic spline interpolation on the signal-to-mask ratios (SMRs) which are obtained from the psychoacoustic model. Experiments on speaker identification and speech re-synthesis showed that the proposed method not only improved the speaker recognition performance, but also improved the speech quality of the re-synthesized speech.

Channel and noise normalization using affine transformed cepstrum

Abstract: The paper addresses the environmental mismatch problem that arises from noise and channel variabilities. A new feature mapping technique based on an optimal affine transform of the cepstrum is proposed to solve the mismatch problem observed over the speaker recognition systems. It is designed based on the fact that both the channel and noise interferences basically cause the cepstrum space to undergo an affine transformation. By taking an inverse transformation, one can easily decouple from the speech the effects of the channel and noise. Alternatively, one can take a forward transform of the training data to simulate the operating conditions.

Chip Design of LPC-cepstrum for Speech Recognition

Abstract: This paper proposed an ASIC of LPC-cepstrum (LPCC) for speech recognition. The proposed ASIC of LPCC can reduce the calculation load of processor in the speech recognition system. In addition, the resource sharing method is adopted into our design in order to reduce the chip size. Hence, it does not give an emphasis on sophistication but on high- performance and low-cost solution. Finally, we did some experiments to compare with other DSP or ASIC design. We found that our proposed LPCC ASIC can efficiently reduce the computation load.

Signal Processing and Feature Extraction

Abstract: The initial stages in speech processing, discussed in Chapter 4, are commonly performed using a short-time Fourier transformation (STFT) of the digitally-sampled acoustic time series. Several representations of the STFT have been employed for automatic speech recognition, including linear, logarithmic scale, logarithmic mel-scale, cepstral and differenced-cepstral coefficients. However, recent investigations of mammalian auditory processing have determined that the cochlea is a time-domain analyzer, and that the STFT representation is not always the most appropriate method of signal analysis. Therefore, this chapter reviews the properties and behavior of cochlear models and their importance to ASR. It emphasizes the benefits gained from better models of “early” signal processing in mammals. A discussion of artificial neural network applications for conventional signal processing problems follows. The remainder of this chapter discusses how low-level “feature maps” may be created and used in ASR applications.