Speech Recognition In Noisy Environments Research Articles

Behavioral assessment of auditory processing disorder in children with non-syndromic cleft lip and/or palate

ObjectivePeripheral hearing disorders have been frequently described in children with non-syndromic cleft lip and/or palate (NSCL/P). However, auditory processing problems are rarely considered for children with NSCL/P despite their poor academic performance in general compared to their craniofacially normal peers. This study aimed to compare auditory processing skills, using behavioral assessment techniques, in school age children with and without NSCL/P. MethodsOne hundred and forty one Mandarin-speaking children with NSCL/P aged from 6.00 to 15.67 years, and 60 age-matched, craniofacially normal children, were recruited. Standard hearing health tests were conducted to evaluate peripheral hearing. Behavioral auditory processing assessment included adaptive tests of temporal resolution (ATTR), and the Mandarin pediatric lexical tone and disyllabic-word picture identification test in noise (MAPPID-N). ResultsAge effects were found in children with cleft disorder but not in the control group for gap detection thresholds with ATTR narrow band noise in the across-channel stimuli condition, with a significant difference in test performance between the 6 to 8 year group and 12 to 15 year group of children with NSCL/P. For MAPPID-N, the bilateral cleft lip and palate subgroup showed significantly poorer SNR-50% scores than the control group in the condition where speech was spatially separated from noise. Also, the cleft palate participants showed a significantly smaller spatial separation advantage for speech recognition in noise compared to the control group children. ConclusionATTR gap detection test results indicated that maturation for temporal resolution abilities was not achieved in children with NSCL/P until approximately 8 years of age compared to approximately 6 years for craniofacially normal children. For speech recognition in noisy environments, poorer abilities to use timing and intensity cues were found in children with cleft palate and children with bilateral cleft lip and palate compared to craniofacially normal children. Consequently, it is worthwhile to consider the potential for auditory processing disorder in when assessing the auditory status of children with NSCL/P.

Functional results after Bonebridge implantation in adults and children with conductive and mixed hearing loss.

In patients with conductive hearing loss caused by middle ear disorders or atresia of the ear canal, a Bonebridge implantation can improve hearing by providing vibratory input to the temporal bone. The expected results are improved puretone thresholds and speech recognition. In the European Union, approval of the Bonebridge implantation was recently extended to children. We evaluated the functional outcome of a Bonebridge implantation for eight adults and three children. We found significant improvement in the puretone thresholds, with improvement in the air-bone gap. Speech recognition after surgery was significantly higher than in the best-aided situation before surgery. The Bonebridge significantly improved speech recognition in noisy environments and sound localization. In situations relevant to daily life, hearing deficits were nearly completely restored with the Bonebridge implantation in both adults and children.

A bio-inspired feature extraction for robust speech recognition

In this paper, a feature extraction method for robust speech recognition in noisy environments is proposed. The proposed method is motivated by a biologically inspired auditory model which simulates the outer/middle ear filtering by a low-pass filter and the spectral behaviour of the cochlea by the Gammachirp auditory filterbank (GcFB). The speech recognition performance of our method is tested on speech signals corrupted by real-world noises. The evaluation results show that the proposed method gives better recognition rates compared to the classic techniques such as Perceptual Linear Prediction (PLP), Linear Predictive Coding (LPC), Linear Prediction Cepstral coefficients (LPCC) and Mel Frequency Cepstral Coefficients (MFCC). The used recognition system is based on the Hidden Markov Models with continuous Gaussian Mixture densities (HMM-GM).

Joint variable frame rate and length analysis for speech recognition under adverse conditions

This paper presents a method that combines variable frame length and rate analysis for speech recognition in noisy environments, together with an investigation of the effect of different frame lengths on speech recognition performance. The method adopts frame selection using an a posteriori signal-to-noise (SNR) ratio weighted energy distance and increases the length of the selected frames, according to the number of non-selected preceding frames. It assigns a higher frame rate and a normal frame length to a rapidly changing and high SNR region of a speech signal, and a lower frame rate and an increased frame length to a steady or low SNR region. The speech recognition results show that the proposed variable frame rate and length method outperforms fixed frame rate and length analysis, as well as standalone variable frame rate analysis in terms of noise-robustness.

Perception of Hearing Aid Users with Nonlinear Frequency Compression on Speech Recognition in Noisy Environment

Introduction: An alternative to ease the descending hearing losses and assist in understanding speech is the use of hearing aids with a nonlinear frequency compression algorithm.

Feature extraction technique using ERB like wavelet sub-band periodic and aperiodic decomposition for TIMIT phoneme recognition

In the recent years, wavelet transform has been found to be an effective tool for the time---frequency analysis for non-stationary and quasi-stationary signals such as speech signals. In the recent past, wavelet transform has been used as feature extraction in speech recognition applications. Here we propose a wavelet based feature extraction technique that signifies both the periodic and aperiodic information along with sub-band instantaneous frequency of speech signal for robust speech recognition in noisy environment. This technique is based on parallel distributed processing technique inspired by the human speech perception process. This frontend feature processing technique employs equivalent rectangular bandwidth (ERB) filter like wavelet speech feature extraction method called Wavelet ERB Sub-band based Periodicity and Aperiodicity Decomposition (WERB-SPADE), and examines its validity for TIMIT phone recognition task in noisy environments. The speech sound is filtered by 24 band ERB like wavelet filter banks, and then the equal loudness pre-emphasized output of each band is processed through comb filter. Each comb filter is designed individually for each frequency sub-band to decompose the signal into periodic and aperiodic features. Thus it takes the advantage of the robustness shown by periodic features without losing certain important information like formant transition incorporated in aperiodic features. Speech recognition experiments with a standard HMM recognizer under both clean-training and multi-training condition training is conducted. Proposed technique shows more robustness compared to other features especially in noisy condition.

Class-Based Parametric Approximation to Histogram Equalization for ASR

This letter assesses an improved equalization transformation for robust speech recognition in noisy environments. The proposal is an evolution of the parametric approximation to Histogram Equalization named PEQ into a two-step algorithm dealing separately with environmental and acoustic mismatch. A first parametric equalization is done to eliminate environmental mismatch. These equalized data are divided into classes, and parametrically re-equalized using class specific references to reduce the acoustic mismatch. Experiments have been conducted for Aurora 2 and Aurora 4 databases. A comparative analysis of the experimental results shows significant benefits for databases with high acoustic variability like Aurora 4.

Real-time spectrum estimation–based dual-channel speech-enhancement algorithm for cochlear implant

BackgroundImprovement of the cochlear implant (CI) front-end signal acquisition is needed to increase speech recognition in noisy environments. To suppress the directional noise, we introduce a speech-enhancement algorithm based on microphone array beamforming and spectral estimation. The experimental results indicate that this method is robust to directional mobile noise and strongly enhances the desired speech, thereby improving the performance of CI devices in a noisy environment.MethodsThe spectrum estimation and the array beamforming methods were combined to suppress the ambient noise. The directivity coefficient was estimated in the noise-only intervals, and was updated to fit for the mobile noise.ResultsThe proposed algorithm was realized in the CI speech strategy. For actual parameters, we use Maxflat filter to obtain fractional sampling points and cepstrum method to differentiate the desired speech frame and the noise frame. The broadband adjustment coefficients were added to compensate the energy loss in the low frequency band.DiscussionsThe approximation of the directivity coefficient is tested and the errors are discussed. We also analyze the algorithm constraint for noise estimation and distortion in CI processing. The performance of the proposed algorithm is analyzed and further be compared with other prevalent methods.ConclusionsThe hardware platform was constructed for the experiments. The speech-enhancement results showed that our algorithm can suppresses the non-stationary noise with high SNR. Excellent performance of the proposed algorithm was obtained in the speech enhancement experiments and mobile testing. And signal distortion results indicate that this algorithm is robust with high SNR improvement and low speech distortion.

Efficient Blind Dereverberation and Echo Cancellation Based on Independent Component Analysis for Actual Acoustic Signals

This letter presents a new algorithm for blind dereverberation and echo cancellation based on independent component analysis (ICA) for actual acoustic signals. We focus on frequency domain ICA (FD-ICA) because its computational cost and speed of learning convergence are sufficiently reasonable for practical applications such as hands-free speech recognition. In applying conventional FD-ICA as a preprocessing of automatic speech recognition in noisy environments, one of the most critical problems is how to cope with reverberations. To extract a clean signal from the reverberant observation, we model the separation process in the short-time Fourier transform domain and apply the multiple input/output inverse-filtering theorem (MINT) to the FD-ICA separation model. A naive implementation of this method is computationally expensive, because its time complexity is the second order of reverberation time. Therefore, the main issue in dereverberation is to reduce the high computational cost of ICA. In this letter, we reduce the computational complexity to the linear order of the reverberation time by using two techniques: (1) a separation model based on the independence of delayed observed signals with MINT and (2) spatial sphering for preprocessing. Experiments show that the computational cost grows in proportion to the linear order of the reverberation time and that our method improves the word correctness of automatic speech recognition by 10 to 20 points in a RT₂₀= 670 ms reverberant environment.

Robust Arabic speech recognition in noisy environments using prosodic features and formant

This paper investigates the contribution of formants and prosodic features such as pitch and energy in Arabic speech recognition under real-life conditions. Our speech recognition system based on Hidden Markov Models (HMMs) is implemented using the HTK Toolkit. The front-end of the system combines features based on conventional Mel-Frequency Cepstral Coefficient (MFFC), prosodic information and formants. The experiments are performed on the ARADIGIT corpus which is a database of Arabic spoken words. The obtained results show that the resulting multivariate feature vectors, in noisy environment, lead to a significant improvement, up to 27%, in word accuracy relative the word accuracy obtained from the state-of-the-art MFCC-based system.

Evaluation of noise reduction technologies in a contemporary cochlear implant system

Evaluation of noise reduction technologies in a contemporary cochlear implant system

Cues above 4 kilohertz can improve spatially separated speech recognition.

There are multiple acoustic cues above 4 kHz that may be used to improve speech recognition in noisy environments. This study tests this hypothesis with spatially separated speech maskers. The reception threshold of speech (RTS) was measured in two conditions: (1) asymmetric condition with target speech at −45° and maskers at +45°, and (2) diffuse condition with speech at 0° and one masker at each of four quadrants around the listener (±45°, ±135°). HINT sentences were rerecorded using a male talker and a 22.05 kHz bandwidth while different males were recorded speaking the television and rainbow as maskers. SRTs were measured with all materials low‐pass filtered at 4, 6, 8, and 10 kHz for the asymmetric and diffuse conditions. 12 normal hearing subjects (on the way to 24) were tested using free‐field loudspeakers in a sound booth. The mean RTSs for the asymmetric condition were −17.43, −19.07, −19.34, and −20.65 dB for 4, 6, 8, and 10 kHz, respectively. For the diffuse condition the mean RTSs were −7.77, −9.00, −9.88, and −10.05 dB for the same frequencies. The results suggest that high‐frequency acoustic cues can enhance the ability to segregate spatially separated speech.

Improving speech intelligibility for hearing aid users using a speech segregation algorithm.

Users of hearing aid devices encounter great difficulty in speech recognition in noisy environments, and show inferior performance in situations with interfering background noise or competing speakers as compared to the case of a quiet background. In such devices, a pre‐processing stage, which first removes the background interference, should greatly boost the speech recognition performance in noisy environments. In this work, the performance of a previously reported speech extraction algorithm is investigated for the purpose of improving the speech intelligibility of hearing aid users. Recently, we performed segregation of speech mixtures using our speech extraction algorithm and showed improvement in the speech recognition performance of normal hearing listeners. In this work, we investigate the performance of our algorithm for users of hearing aids on the same perceptual task of recognizing speech in the presence of either interfering speech or noise. We focus on tuning the segregation algorithm for hearing aid users, and emphasize on the preservation of the appropriate envelope and fine structure cues which will prove most helpful in the recognition of the target speech signal.

Application of shape analysis techniques for improved CASA-based speech separation

This paper addresses the problem of single channel speech separation for robust speech recognition in noisy environments. We propose to apply shape analysis techniques in image analysis to a computational auditory scene analysis (CASA)-based speech separation system. The CASA-based system extracts the desired speech signals from a mixture of speech signals and other sound sources. In the proposed method, we complement the missing speech signals by applying the shape analysis techniques such as labeling and distance function. In the speech separation experiment using various artificial mixtures of speech and various noise sources, the proposed system increases the signal-to-noise ratio by 6.6 dB. In the speech recognition experiment using the Interspeech speech database, it improves recognition accuracy by 22% for recognition of speech added with speech-shaped stationary noise.

Improvement of the speech recognition in noisy environments using a nonparametric regression

In this paper, an efficient speech recognition system based on the general regression neural network (GRNN) has been presented. The GRNN has been previously applied for phoneme identification and isolated word recognition in quiet environment. We propose to extend this method to Arabic spoken word recognition in adverse conditions because noise robustness is one of the most challenging problems in automatic speech recognition (ASR). The proposed system has been tested for Arabic digit recognition at different signal-to-noise ratio (SNR) levels in various noisy conditions, including stationary and nonstationary background noises issued from NOISEX-92 database. The proposed scheme is compared with the similar recognisers based on the multilayer perceptron (MLP), the Elman recurrent neural network (RNN) and the discrete hidden Markov model (HMM). The experimental results show that the use of the neural network approach including nonparametric regression improves the global performance of the speech recogniser in noisy environments.

Incorporating the voicing information into HMM-based automatic speech recognition in noisy environments

In this paper, we propose a model for the incorporation of voicing information into a speech recognition system in noisy environments. The employed voicing information is estimated by a novel method that can provide this information for each filter-bank channel and does not require information about the fundamental frequency. The voicing information is modelled by employing the Bernoulli distribution. The voicing model is obtained for each HMM state and mixture by a Viterbi-style training procedure. The proposed voicing incorporation is evaluated both within a standard model and two other models that had compensated for the noise effect, the missing-feature and the multi-conditional training model. Experiments are first performed on noisy speech data from the Aurora 2 database. Significant performance improvements are achieved when the voicing information is incorporated within the standard model as well as the noise-compensated models. The employment of voicing information is also demonstrated on a phoneme recognition task on the noise-corrupted TIMIT database and considerable improvements are observed.

The Importance of Auricular Prostheses for Speech Recognition

To examine the effects of an auricular prosthesis on sound levels at the entrance of the ear canal by measuring the auricular prosthesis transfer function (APTF) and to determine the effect of the prosthesis on speech recognition in noisy hearing conditions. Eight prostheses were used to measure the APTF. A microphone at the entrance of the ear canal measured sound pressure levels with the prosthesis present or absent while the head was rotated 360 degrees at 30 degrees increments. The Hearing in Noise Test was modified by the APTF to simulate the absence of an auricular prosthesis. Speech recognition was measured by testing 11 subjects with the unmodified Hearing in Noise Test and the modified Hearing in Noise Test. The APTF changed with the head's position relative to the speaker. The mean (SD) maximal gain provided by an auricular prosthesis was 8.1 (2.7) dB at 4.6 (1.0) kHz and 9.7 (1.7) dB at 11.5 (0.9) kHz at 0 degrees rotation. During speech testing, the auricular prosthesis improved the mean (SD) signal to noise ratio by 1.7 (1.7) dB at 0 degrees (P< .001), 0.9 (2.2) dB at 90 degrees (P=.04), and 0.5 (2.3) dB at 180 degrees (P=.52). The acoustic gain provided by an auricular prosthesis increases speech recognition in noisy environments. Auricular prostheses not only restore aesthetics but also improve hearing.

R442 – Antioxidant Therapy Prevents Presbycusis In C57BL6 mice

Problem Presbycusis, or age-related hearing loss, is characterized by gradual, progressive sensorineural hearing loss which accompanies aging with associated decreased speech recognition in noisy environments, slowed central processing of acoustic stimuli and impaired sound localization. In addition to impairing one's ability to communicate effectively, presbycusis jeopardizes one's autonomy, presents a safety concern and has been correlated with an increased incidence of clinical depression and social withdrawal. By 2025, it is estimated that approximately 25 million Americans will be affected, thus this presents a major public health concern. The purpose of this study is to evaluate the potential of antioxidant therapy, including two novel antioxidants, in the prevention of presbycusis. Methods C57BL6 mice were assigned to treatment or control groups. Treatment groups of mice were fed with a combination of two novel antioxidants, L-cysteine-glutathione mixed disulfide and ribose cysteine, as well as vitamin B12, folic acid, and ascorbic acid. Auditory brainstem response thresholds were recorded at baseline and on a monthly basis following initiation of treatment. Results Threshold shifts were observed according to the established pattern of age related loss associated with C57BL6 mice. Threshold shifts were decreased in the treatment group. Statistically significant differences (p&lt;0.001) in threshold shifts were observed between the treatment and control groups for click stimulus and pure tones at 4, 8, 12, 16 and 32 kHz. The treatment group was found to have a smaller degree of shift in association with aging. Conclusion Combination antioxidant therapy effectively decreased threshold shifts on auditory brainstem responses within the established animal model for presbycusis. Significance Combination antioxidant therapy may prove a safe and cost-effective method of preventing presbycusis in our growing elderly population. Support American Academy of Otolaryngology – Head and Neck Surgery Foundation Resident Research Grant.

Perceptual features for automatic speech recognition in noisy environments

The performances of two perceptual properties of the peripheral auditory system, synaptic adaptation and two-tone suppression, are compared for automatic speech recognition (ASR) in an additive noise environment. A simple method of synaptic adaptation as determined by psychoacoustic observations was implemented with temporal processing of speech utilizing a zero-crossing auditory model as a pre-processing front end. The concept is similar to RASTA processing, but instead of bandpass filters, a high-pass infinite impulse response (IIR) filter is used. It is shown that rapid synaptic adaptation may be implemented by temporal processing using the zero-crossing algorithm, not otherwise implementable in the spectral domain implementation. The two-tone suppression was implemented in the zero-crossing auditory model using a companding strategy. Recognition performances with the two perceptual features were evaluated on isolated digits (TIDIGITS) corpus using continuous density HMM recognizer in white, factory, babble and Volvo noise. It is observed that synaptic adaptation performs better in stationary white Gaussian noise. In presence of non-stationary non-Gaussian noise, however, no improvements or a degradation is observed. Moreover, a reciprocal effect is observed with two-tone suppression, with better performance in non-Gaussian real-world noise and degradation in stationary white Gaussian noise.

An efficient frame selection approach to variable frame rate analysis for noise robust speech recognition

This paper presents a low-complexity, effective variable frame rate (VFR) analysis method that conducts frame selection on the basis of a posteriori signal-to-noise ratio (SNR) weighted energy distance. It has two characteristics. First, energy distance (instead of cepstral distance) is used to make it computationally efficient and thus enable a finer granularity in search as compared with cepstral distance criterion. Secondly, SNR weighting is used to emphasize the reliable regions in noisy speech signals. In terms of frame selection, it is experimentally found that the method is able to assign a higher frame rate to fast changing events such as consonants, a lower frame rate to steady regions like vowels and no frames to silence, even for very low SNR signals. The VFR method is applied to speech recognition in noisy environments to improve noise robustness. Being a method that takes effect in the time-domain, it is moreover combined with spectral- and cepstral-domain techniques to gain further improvement. Experiments are conducted on the Aurora 2 database, which is the TI digits database artificially distorted by adding different noises, and very encouraging results are obtained.