Speech Recognition In Noisy Conditions Research Articles

Efficacy of Bone-Anchored Hearing Aids in Single-Sided Deafness: A Systematic Review.

Bone-anchored hearing aids (BAHAs) have been known to partially restore some of the functions lost in subjects with single-sided deafness (SSD). Our aims in this systemic review were to analyze the present capabilities of BAHAs in the context of SSD, and to evaluate the efficacy of BAHAs in improving speech recognition in noisy conditions, sound localization, and subjective outcomes. A systematic search was undertaken until August 2015 by two independent reviewers, with disagreements resolved by consensus. Among 286 references, we analyzed 14 studies that used both subjective and objective indicators to assess the capabilities of a total of 296 patients in the unaided and aided situations. Although there was "no benefit" of BAHA implantation for sound localization, BAHAs certainly improved subjects' speech discrimination in noisy circumstances. In the six studies that dealt with sound localization, no significant difference was found after the implantation. Twelve studies showed the benefits of BAHAs for speech discrimination in noise. Regarding subjective outcomes of using the prosthesis in patients with SSD (abbreviated profile of hearing aid benefit [APHAB] and the Glasgow hearing aid benefit profile [GHABP], etc.), we noticed an improvement in the quality of life. This systematic review has indicated that BAHAs may successfully rehabilitate patients with SSD by alleviating the hearing handicap to a certain degree, which could improve patients' quality of life. This report has presented additional evidence of effective auditory rehabilitation for SSD and will be helpful to clinicians counseling patients regarding treatment options for SSD.

Noise-Robust Speech Recognition Through Auditory Feature Detection and Spike Sequence Decoding

Speech recognition in noisy conditions is a major challenge for computer systems, but the human brain performs it routinely and accurately. Automatic speech recognition (ASR) systems that are inspired by neuroscience can potentially bridge the performance gap between humans and machines. We present a system for noise-robust isolated word recognition that works by decoding sequences of spikes from a population of simulated auditory feature-detecting neurons. Each neuron is trained to respond selectively to a brief spectrotemporal pattern, or feature, drawn from the simulated auditory nerve response to speech. The neural population conveys the time-dependent structure of a sound by its sequence of spikes. We compare two methods for decoding the spike sequences--one using a hidden Markov model-based recognizer, the other using a novel template-based recognition scheme. In the latter case, words are recognized by comparing their spike sequences to template sequences obtained from clean training data, using a similarity measure based on the length of the longest common sub-sequence. Using isolated spoken digits from the AURORA-2 database, we show that our combined system outperforms a state-of-the-art robust speech recognizer at low signal-to-noise ratios. Both the spike-based encoding scheme and the template-based decoding offer gains in noise robustness over traditional speech recognition methods. Our system highlights potential advantages of spike-based acoustic coding and provides a biologically motivated framework for robust ASR development.

On the Use of Asymmetric Windows for Robust Speech Recognition

This paper deals with the problem of searching for a suitable window for robust speech recognition in noisy conditions. A set of asymmetric windows, so-called DDRc,w, are proposed which are controlled by two parameters, center c and width w. These windows are derived from the DDR window used in the higher-lag autocorrelation spectrum estimation (HASE) method and act over the OSA (One-Sided Autocorrelation) in order to perform spectral estimation. The two parameters, c and w, allow us to control the level of weight given to the first noisy autocorrelation coefficients and to emphasize the important ones. Finally, it is shown that the best window of the proposed set is the DDR62,200. This window is centered around the average pitch of human speech and it provides a higher speech recognition performance over the Aurora-2 and Aurora-3 databases than those obtained by previously proposed windows.

On the Effects of Filterbank Design and Energy Computation on Robust Speech Recognition

In this paper, we examine how energy computation and filterbank design contribute to the overall front-end robustness, especially when the investigated features are applied to noisy speech signals, in mismatched training-testing conditions. In prior work (“Auditory Teager energy cepstrum coefficients for robust speech recognition,” D. Dimitriadis, P. Maragos, and A. Potamianos, in Proc. Eurospeech'05, Sep. 2005), a novel feature set called “Teager energy cepstrum coefficients” (TECCs) has been proposed, employing a dense, smooth filterbank and alternative energy computation schemes. TECCs were shown to be more robust to noise and exhibit improved performance compared to the widely used Mel frequency cepstral coefficients (MFCCs). In this paper, we attempt to interpret these results using a combined theoretical and experimental analysis framework. Specifically, we investigate in detail the connection between the filterbank design, i.e., the filter shape and bandwidth, the energy estimation scheme and the automatic speech recognition (ASR) performance under a variety of additive and/or convolutional noise conditions. For this purpose: 1) the performance of filterbanks using triangular, Gabor, and Gammatone filters with various bandwidths and filter positions are examined under different noisy speech recognition tasks, and 2) the squared amplitude and Teager-Kaiser energy operators are compared as two alternative approaches of computing the signal energy. Our end-goal is to understand how to select the most efficient filterbank and energy computation scheme that are maximally robust under both clean and noisy recording conditions. Theoretical and experimental results show that: 1) the filter bandwidth is one of the most important factors affecting speech recognition performance in noise, while the shape of the filter is of secondary importance, and 2) the Teager-Kaiser operator outperforms (on the average and for most noise types) the squared amplitude energy computation scheme for speech recognition in noisy conditions, especially, for large filter bandwidths. Experimental results show that selecting the appropriate filterbank and energy computation scheme can lead to significant error rate reduction over both MFCC and perceptual linear predicion (PLP) features for a variety of speech recognition tasks. A relative error rate reduction of up to ~ 30% for MFCCs and ~ 39% for PLPs is shown for the Aurora-3 Spanish Task.

Investigating the effects of stimulus duration and context on pitch perception by cochlear implant users

Cochlear implant sound processing strategies that use time-varying pulse rates to transmit fine structure information are one proposed method for improving the spectral representation of a sound with the eventual goal of improving speech recognition in noisy conditions, speech recognition in tonal languages, and music identification and appreciation. However, many of the perceptual phenomena associated with time-varying rates are not well understood. In this study, the effects of stimulus duration on both the place and rate-pitch percepts were investigated via psychophysical experiments. Four Nucleus CI24 cochlear implant users participated in these experiments, which included a short-duration pitch ranking task and three adaptive pulse rate discrimination tasks. When duration was fixed from trial-to-trial and rate was varied adaptively, results suggested that both the place-pitch and rate-pitch percepts may be independent of duration for durations above 10 and 20 ms, respectively. When duration was varied and pulse rates were fixed, performance was highly variable within and across subjects. Implications for multi-rate sound processing strategies are discussed.

Informational masking in first‐ and second‐language speech recognition.

Human speech recognition in noisy conditions is modulated by cognitive factors such as language background. For example, noise is more detrimental to non‐native listeners than native listeners (e.g., van Wijngaarden et al., 2002), and when noise is a speech signal, native‐language noise is more detrimental than foreign‐language noise for listeners attending to native‐language speech targets (e.g., Van Engen and Bradlow, 2007). It is not clear, however, whether this increased interference is primarily due to the native status of the noise or to the greater similarity between target and noise. To address this issue, English speech recognition in the presence of English and Mandarin babble was assessed for monolingual English listeners and L2 English listeners whose L1 is Mandarin. Results showed that intelligibility for both groups was lower in English versus Mandarin babble; that is, L2 listeners experienced more difficulty in same‐language noise versus native‐language noise. However, monolingual English listeners showed a greater release from masking in Mandarin noise than did L1 Mandarin listeners. The informational masking imposed on speech targets by interfering speech noise, therefore, is dependent both on the linguistic and/or acoustic similarity between the noise and the target and on the listeners’ experience with the languages involved.

Features based on filtering and spectral peaks in autocorrelation domain for robust speech recognition

In this paper, a set of features derived by filtering and spectral peak extraction in autocorrelation domain are proposed. We focus on the effect of the additive noise on speech recognition. Assuming that the channel characteristics and additive noises are stationary, these new features improve the robustness of speech recognition in noisy conditions. In this approach, initially, the autocorrelation sequence of a speech signal frame is computed. Filtering of the autocorrelation of speech signal is carried out in the second step, and then, the short-time power spectrum of speech is obtained from the speech signal through the fast Fourier transform. The power spectrum peaks are then calculated by differentiating the power spectrum with respect to frequency. The magnitudes of these peaks are then projected onto the mel-scale and pass the filter bank. Finally, a set of cepstral coefficients are derived from the outputs of the filter bank. The effectiveness of the new features for speech recognition in noisy conditions will be shown in this paper through a number of speech recognition experiments. A task of multi-speaker isolated-word recognition and another one of multi-speaker continuous speech recognition with various artificially added noises such as factory, babble, car and F16 were used in these experiments. Also, a set of experiments were carried out on Aurora 2 task. Experimental results show significant improvements under noisy conditions in comparison to the results obtained using traditional feature extraction methods. We have also reported the results obtained by applying cepstral mean normalization on the methods to get robust features against both additive noise and channel distortion.

Introducing complexity measures in nonlinear physiological signals: application to robust speech recognition

Information measures have been used in the context of nonlinear systems presenting abrupt complexity changes and related to nonlinear time series analysis. In this study, complexity measures such as Shannon entropy, q-entropy and their associated divergences have been added to a robust speech recognizer front-end. The method proposed here is tested on continuous speech and compared with a classical mel-cepstral analysis. The recognition degradation has been evaluated in both systems in presence of white and babble noise. The results suggest that complexity measures provide additional valuable information for speech recognition in noisy conditions.