This paper presents robust speech enhancement using noise estimation based on smoothing of spectral noise floor (SNF) for nonstationary noise environments. The spectral gain function is obtained by well-known log-spectral amplitude (LSA) estimation criterion associated with the speech presence uncertainty. The noise estimate is given by averaging actual spectral power values, using a smoothing ...

Supervised masking approaches in the time-frequency domain aim to employ deep neural networks estimate a multiplicative mask extract clean speech. This leads single for each input without any guarantees or measures of reliability. In this paper, we study benefits modeling uncertainty speech estimation. Prediction is typically categorized into aleatoric and epistemic uncertainty. The former refe...

Background noise can have a significant impact on the performance of speech recognition systems. A range of fast featurespace and model-based schemes have been investigated to increase robustness. Model-based approaches typically achieve lower error rates, but at an increased computational load compared to feature-based approaches. Thismakes their use inmany situations impractical. The uncertai...

background: according to previous studies, most of the speech recognition disorders in older adults are the results of deficits in audibility and auditory temporal resolution. in this paper, the effect of ageing on timecompressed speech and auditory temporal resolution by word recognition in continuous and interrupted noise was studied. methods: a time-compressed speech test (tcst) was conducte...

This paper presents a new hidden Markov model-based (HMM-based) speech enhancement framework based on the independent component analysis (ICA). We propose analytical procedures for training clean speech and noise models by the Baum re-estimation algorithm and present a Maximum a posterior (MAP) estimator based on Laplace-Gaussian (for clean speech and noise respectively) combination in the HMM ...

Good public speaking skills are essential in many professions as well as everyday life, but speech anxiety is a common problem. While it is established that public speaking training in virtual reality (VR) is effective, comprehensive studies on the underlying factors that contribute to this success are rare. The “quality evaluation of user-system interaction in virtual reality” framework for ev...

 Background: According to previous studies, most of the speech recognition disorders in older adults are the results of deficits in audibility and auditory temporal resolution. In this paper, the effect of ageing on timecompressed speech and auditory temporal resolution by word recognition in continuous and interrupted noise was studied. Methods: A time-compressed speech test (TCST) w...