Abstract:This paper addresses the problem of single microphone speech enhancement in noisy environments. Common short-time noise reduction techniques proposed in the art are expressed as a spectral gain depending on the a priori SNR. In the well-known decisiondirected approach, the a priori SNR depends on the speech spectrum estimation in the previous frame. As a consequence the gain function matches the previous frame rather than the current one which degrades the noise reduction performance. We propose a new method c… Show more
“…This is further explored by Cohen (2005), and is used in a modified form in ETSI (2002); Plapous et al (2004). Whilst these approaches are beyond the scope of the present study, the proposed approach does not preclude using them.…”
Cepstral normalisation in automatic speech recognition is investigated in the context of robustness to additive noise. In this paper, it is argued that such normalisation leads naturally to a speech feature based on signal to noise ratio rather than absolute energy (or power). Explicit calculation of this SNR-cepstrum by means of a noise estimate is shown to have theoretical and practical advantages over the usual (energy based) cepstrum. The relationship between the SNR-cepstrum and the articulation index, known in psycho-acoustics, is discussed. Experiments are presented suggesting that the combination of the SNR-cepstrum with the well known perceptual linear prediction method can be beneficial in noisy environments.
“…This is further explored by Cohen (2005), and is used in a modified form in ETSI (2002); Plapous et al (2004). Whilst these approaches are beyond the scope of the present study, the proposed approach does not preclude using them.…”
Cepstral normalisation in automatic speech recognition is investigated in the context of robustness to additive noise. In this paper, it is argued that such normalisation leads naturally to a speech feature based on signal to noise ratio rather than absolute energy (or power). Explicit calculation of this SNR-cepstrum by means of a noise estimate is shown to have theoretical and practical advantages over the usual (energy based) cepstrum. The relationship between the SNR-cepstrum and the articulation index, known in psycho-acoustics, is discussed. Experiments are presented suggesting that the combination of the SNR-cepstrum with the well known perceptual linear prediction method can be beneficial in noisy environments.
“…We note that in the Decision Directed estimator of [3], the ML estimate of ξ of (13) is regularised using an estimate based on the previous spectral magnitude estimate. This is further explored by Cohen [10], and is used in a modified form in [4,11]. Whilst these approaches are beyond the scope of the present study, our approach does not preclude using them.…”
When combined with cepstral normalisation techniques, the features normally used in Automatic Speech Recognition are based on Signal to Noise Ratio (SNR). We show that calculating SNR from the outset, rather than relying on cepstral normalisation to produce it, gives features with a number of practical and mathematical advantages over power-spectral based ones. In a detailed analysis, we derive Maximum Likelihood and Maximum a-Posteriori estimates for SNR based features, and show that they can outperform more conventional ones, especially when subsequently combined with cepstral variance normalisation. We further show anecdotal evidence that SNR based features lend themselves well to noise estimates based on low-energy envelope tracking.
“…A several one-, two-and multichannel sensors techniques are proposed to deal with this problem. For example in [14][15][16][17], several single and two-sensor techniques are proposed to correct these distortions [18][19][20]. More advanced techniques are then proposed recently in [21,22].…”
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.