Recent literature indicates increasing interest in deep neural networks for use in speech enhancement systems. Currently, these systems are mostly evaluated through objective measures of speech quality and/or intelligibility. Subjective intelligibility evaluations of these systems have so far not been reported. In this paper we report the results of a speech recognition test with 15 participant...

When interacting with other people, we assume that they have their reasons for what do and believe, experience recognizable feelings emotions. people act from weakness of will or are otherwise irrational, can still be comprehensible to us, since know it is like fall temptation against one’s better judgment. Still, when someone’s experiences, way thinking vastly different our own, understanding ...

The aim of the research described in this paperwas to ascertain whether or not there are differences in intelligibility between male and female voices under noise conditions which are spectrally in agreement with everyday speech maskers. An experiment was carried out in which the CVC and phoneme intelligibility of ten male and female voices was measured. The listeners also gave their opinion on...

In this paper, we propose an algorithm which enhances the speech intelligibility using the properties of human auditory system. In previous algorithms related to the speech intelligibility, the improvement in intelligibility has been mostly incorporated in a single-channel environment where the speech and noise signals are mixed together. But the speech enhancement problem of dual channel, in w...

The alteration in the pitch of speech produced by breathing helium mixtures suggests the possibility of increased intelligibility of such speech by selective frequency filtering. This study evaluated speech, spoken after breathing an 80 % — 20 % mixture which was combined with broadband noise and then passed through a selection of bandpass filters. Results indicate that while intelligibility di...

Speakers appear to adopt strategies to improve speech intelligibility for interlocutors in adverse acoustic conditions. Generated speech, whether synthetic, recorded or live, may also benefit from context-sensitive modifications in challenging situations. The current study measured the effect on intelligibility of six spectral and temporal modifications operating under global constraints of con...

This review article provides an overview of differences between conversational (or cnv) and clear (or clr) speech, for a variety of speakers, in terms of speech intelligibility, and in terms of acoustic characteristics. Researchers have studied the relationship between acoustic features and speech intelligibility by, for example, studying correlations. However, the question “which acoustic feat...

The objective intelligibility assessment of nonlinearly enhanced speech is a widely experienced problem. Nonlinear speech enhancement processors operate primarily on the low-level and transient components of speech. As these sections contain important acoustic cues as well as context-constitutive information, they dominate speech intelligibility. For that reason, shorttime intelligibility measu...

Intelligibility in a telecommunication-based conference application can largely be improved, when the speech signals of the different interlocutors are presented spatially, and when the sources are spatially distributed. For the controlled exploitation of the underlying Cocktail Party Effect in speech chat applications, a prediction of the intelligibility related to a given spatial configuratio...

In this paper our aim is to investigate the properties of the modulation domain and more specifically, to evaluate the relative contributions of the modulation magnitude and phase spectra towards speech intelligibility. For this purpose, we extend the traditional (acoustic domain) analysis–modification–synthesis framework to include modulation domain processing. We use this framework to constru...