Speech quality and intelligibility are of significant importance during clinical hearing aid (HA) fitting and verification. Validated intrusive objective predictors of intelligibility and quality such as the Hearing Aid Speech Perception Index (HASPI) and the Hearing Aid Speech Quality Index (HASQI) have not been widely adopted for implementation within clinically available HA test systems. Recent advances in non-intrusive measures, such as those from Clarity Prediction Challenges (CPCs) and HASA-Net, are also not yet accessible to clinicians. Moreover, most of these advancements rely on datasets from simulated HAs, not the commercial devices used by audiologists. This work aims to develop non-intrusive quality and intelligibility indices using custom databases of noisy speech recorded in a HA test box. Similar to the successful models from previous CPCs, the indices were derived using a novel non-intrusive model leveraging automatic speech recognition and self-supervised learning techniques. The proposed non-intrusive model was trained to predict the intrusive HASPI/HASQI values and later validated against subjective intelligibility data obtained from a group of listeners with hearing loss. The proposed model resulted in strong correlations with HASQI (95.96%) and HASPI (97.59%), and a moderate correlation with the subjective intelligibility scores (76.63%).

Objective measures of speech quality are highly desirable in benchmarking and monitoring the performance of hearing aids (HAs). Existing HA speech quality indices such as the hearing aid speech quality index (HASQI) are intrusive in that they require a properly time-aligned and frequency-shaped reference signal to predict the quality of HA output. Two new reference-free HA speech quality indices are proposed in this paper, based on a model that amalgamates perceptual linear prediction (PLP), hearing loss (HL) modeling, and machine learning concepts. For the first index, HL-modified PLP coefficients and their statistics were used as the feature set, which was subsequently mapped to the predicted quality scores using support vector regression (SVR). For the second index, HL-impacted gammatone auditory filterbank energies and their second-order statistics constituted the feature set, which was again mapped using SVR. Two databases involving HA recordings were collected and utilized for the evaluation of the robustness and generalizability of the two indices. Experimental results showed that the index based on the gammatone filterbank energies not only correlated well with HA quality ratings by hearing impaired listeners, but also exhibited robust performance across different test conditions and was comparable to the full-reference HASQI performance.

A non-intrusive objective speech quality and intelligibility measure tailored to hearing restoration instruments is proposed and evaluated in complex listening environments. The measure builds upon the previously-proposed “speech-to-reverberation modulation energy ratio” (SRMR) by incorporating hearing impairment percepts, such as hearing loss thresholds and altered modulation frequency selectivity. Performance is assessed using speech data corrupted by additive noise, reverberation, and noise-plus-reverberation which were subjectively rated by cochlear implant and hearing aid users. Experimental results show that the developed measures outperform the original SRMR metric for hearing impaired listeners and achieve performance levels inline with existing intrusive quality and intelligibility metrics, but with the advantage of not requiring access to a clean reference signal. As such, the measure may be used to develop qualityor intelligibility-aware speech enhancement algorithms for advanced hearing restoration instruments.

In speech enhancement applications, a validated metric of noise reduction performance is vital in the relative ranking of noise reduction algorithms and in enhancing the performance of a noise reduction algorithm. subjective scores of enhanced speech remain the yardstick for performance, but objective metrics that emulate subjective evaluations are preferred for cost- and time-effectiveness. In this paper, we analyze the performance of two objective methods for predicting the quality of enhanced speech. The first method employs the coherence-based speech intelligibility index, while the second method uses features derived from the Moore-Glasberg auditory model. In both cases, the features are mapped to a quality score using the Bayesian modeling approach. Results show that the combination of the auditory model-based feature set and the Bayesian modeling provides the best performance in predicting the quality scores of enhanced speech.

Total laryngectomy is often the treatment of choice for many patients suffering from laryngeal cancer. This procedure alters the speech production mechanism, and tracheoesophageal (TE) speech is an alternative where the pulmonary air is forced through the esophagus. TE speech is often characterized by poor intelligibility and voice quality. Acoustic analysis of TE speech has the potential of quantifying the voice quality and assisting the speech pathologist in determining and monitoring the therapy process. In this paper, we apply two different methods for predicting the voice quality ratings of TE speakers by naive listeners: (a) conventional spectral and linear prediction measurements that were investigated in earlier studies, and (b) a methodology based on a perceptual auditory model that attempts to mimic the speech quality perception by a normal hearing listener. Experimental results with a database of 35 TE speakers showed that the auditory-model based approach significantly outperforms the traditional methods.

A novel Bayesian model-based non-intrusive speech quality evaluation (BM-NiSQE) algorithm is presented in this paper. The proposed BM-NiSQE algorithm employs a statistical model approach and Bayesian inference to estimate the speech quality only using the output signal of the system under test. In the proposed algorithm, the speech features are extracted by perceptual spectral analysis. Gaussian mixture density hidden Markov models (GMD-HMMs) are exploited to characterize different speech quality categories, which take into account not only the temporal variations of speech signal but also the spectral statistical characteristics in the perception domain. Based on the trained GMD-HMMs, the prediction of speech quality is carried out by Bayesian inference and minimum mean square error (MMSE) estimation. Preliminary experimental results show that the predicted results of the proposed algorithm correlate well with the subjective quality scores.

The current generation of digital hearing aids perform amplitude compression in multiple channels. The resulting non-linear processing has the potential to create distortion components which reduce their effectiveness and user satisfaction. The performance of these devices is typically evaluated using subjective test procedures. While these approaches are preferred, they are time consuming. Objective electroacoustic measurements of speech quality are attractive but require effective modeling of these devices. Subband adaptive modeling architectures have been found to facilitate these measurements. It is shown that optimal modeling occurs only when the number of channels in the subband adaptive model matches the number of hearing aid channels. A technique based on combined sinusoidal and broadband noise excitation is then exploited to identify the number of channels in the hearing aid to be used in the optimal subband adaptive model.

This letter presents a novel nonintrusive speech quality evaluation method using an adaptive neurofuzzy inference system (ANFIS). The proposed method employed a first-order Sugeno-type fuzzy inference system (FIS) to estimate speech quality using only the output signal of the system under test. This new method was compared with the state-of-the-art nonintrusive quality evaluation standard, the ITU-T P.563 Recommendation, using seven subjective quality databases of the ITU-T P-series Supplementary 23. Experimental results show that the correlation of the proposed method with the subjective quality scores reached 0.8812, with a standard error of 0.3647 across the entire database. This compares favorably with the standard P.563, which provides a correlation and standard error of 0.8422 and 0.4493, respectively.

A new hidden Markov model (HMM) based frequency bandwidth extension algorithm using line spectral frequencies (HMM-LSF-FBE) is proposed. The proposed algorithm improves the performance of the traditional LSF-based extension algorithm by exploiting an HMM to indicate the proper representatives of different speech frames, and by applying a minimum mean square criterion to estimate the high-band LSF values. The proposed algorithm has been tested and compared to the traditional LSF-based algorithm in terms of the perceptual evaluation of speech quality (PESQ) objective measure and speech spectrograms. Simulation results show that the proposed algorithm outperforms the traditional method by eliminating undesired whistling sounds completely. In addition, the bandwidth extended speech signals created by the proposed algorithm are significantly more pleasant to the human ear than the original narrowband speech signals from which they are derived.

Adaptive modeling of digital hearing aids is useful in characterizing the hearing aid behavior in response to “real world” stimuli such as speech and music. Most modern hearing aids employ amplitude compression in different frequency bands for effective mapping of the wide dynamic range audio signals into the reduced dynamic range of the hearing impaired listeners. Due to the presence of independent compression channels, the conventional fullband adaptive model might not adequately characterize the performance of a multichannel compression hearing aid (MCHA). In this paper, we propose a subband adaptive modeling approach to characterize the electroacoustic performance of a MCHA. The proposed structure employs uniform, oversampled DFT filterbanks for analysis and synthesis, and the affine projection algorithm for adaptive modeling in each subband. Experiments with simulated MCHAs showed that the subband structure outperforms the fullband structure under a variety of operating conditions.

Acoustic measures of vocal function are routinely used for the assessment of disordered voice, and for monitoring patient's progress over the course of therapy. In current clinical practice, acoustic measures extracted from sustained vowels are used for vocal function characterization. However, the measures derived from continuous speech samples are required for accurate assessment of voice quality. In this paper, a time-frequency approach for pathological voice discrimination has been proposed. The speech signals were decomposed using an adaptive time-frequency transform algorithm, and the signal decomposition parameters such as the octave (scale) maximum, octave mean, energy rate, and length ratio were analyzed using the maximum likelihood method and Jack-knife algorithm for classification. A classification accuracy of 90% was obtained with a database of 40 speech signals (20 normal and 20 pathological cases).

Somatosensory evoked potentials (SEPs) are extremely useful in peripheral nerve monitoring and in the diagnosis of various neuromuscular disorders. However, surface measurements of these potentials often result in imperceptible SEP waveforms due to the very poor signal-to-noise ratio (SNR). Adaptive noise cancelling is an attractive technique which can be used to improve this poor SNR. One of the important factors that affect the performance of an adaptive noise canceller (ANC) is the presence of SEP components in the reference channel of the ANC. The authors propose a novel multichannel crosstalk resistant adaptive noise canceller (MCRANC) for offsetting the problems caused by the "SEP" crosstalk. The performance of this MCRANC is evaluated analytically and through simulations.