Enhancing Auditory Selective Attention Using a Visually Guided Hearing Aid

Practical utility of a head-mounted gaze-directed beamforming system

Assistive auditory devices that enhance signal-to-noise ratio must follow the user's changing attention; errors could lead to the desired source being suppressed as noise. A method for measuring the practical benefit of attention-following speech enhancement is described and used to show a benefit for gaze-directed beamforming over natural binaural hearing. First, participants watched a recorded video conference call between two people with six additional interfering voices in different directions. The directions of the target voices corresponded to the spatial layout of their video streams. A simulated beamformer was yoked to the participant's gaze direction using an eye tracker. For the control condition, all eight voices were spatially distributed in a simulation of unaided binaural hearing. Participants completed questionnaires on the content of the conversation, scoring twice as high in the questionnaires for the beamforming condition. Sentence-by-sentence intelligibility was then measured using new participants who viewed the same audiovisual stimulus for each isolated sentence. Participants recognized twice as many words in the beamforming condition. The results demonstrate the potential practical benefit of gaze-directed beamforming for hearing aids and illustrate how detailed intelligibility data can be retrieved from an experiment that involves behavioral engagement in an ongoing listening task.

Multilevel Modeling of Gaze From Listeners With Hearing Loss Following a Realistic Conversation

PURPOSE

There is a need for tools to study real-world communication abilities in people with hearing loss. We outline a potential method for this that analyzes gaze and use it to answer the question of when and how much listeners with hearing loss look toward a new talker in a conversation.

METHOD

Twenty-two older adults with hearing loss followed a prerecorded two-person audiovisual conversation in the presence of babble noise. We compared their eye-gaze direction to the conversation in two multilevel logistic regression (MLR) analyses. First, we split the conversation into events classified by the number of active talkers within a turn or a transition, and we tested if these predicted the listener's gaze. Second, we mapped the odds that a listener gazed toward a new talker over time during a conversation transition.

RESULTS

We found no evidence that our conversation events predicted changes in the listener's gaze, but the listener's gaze toward the new talker during a silence-transition was predicted by time: The odds of looking at the new talker increased in an s-shaped curve from at least 0.4 s before to 1 s after the onset of the new talker's speech. A comparison of models with different random effects indicated that more variance was explained by differences between individual conversation events than by differences between individual listeners.

CONCLUSIONS

MLR modeling of eye-gaze during talker transitions is a promising approach to study a listener's perception of realistic conversation. Our experience provides insight to guide future research with this method.

Individual differences in speech-on-speech masking are correlated with cognitive and visual task performance

Individual differences in spatial tuning for masked target speech identification were determined using maskers that varied in type and proximity to the target source. The maskers were chosen to produce three strengths of informational masking (IM): high [same-gender, speech-on-speech (SOS) masking], intermediate (the same masker speech time-reversed), and low (speech-shaped, speech-envelope-modulated noise). Typical for this task, individual differences increased as IM increased, while overall performance decreased. To determine the extent to which auditory performance might generalize to another sensory modality, a comparison visual task was also implemented. Visual search time was measured for identifying a cued object among "clouds" of distractors that were varied symmetrically in proximity to the target. The visual maskers also were chosen to produce three strengths of an analog of IM based on feature similarities between the target and maskers. Significant correlations were found for overall auditory and visual task performance, and both of these measures were correlated with an index of general cognitive reasoning. Overall, the findings provide qualified support for the proposition that the ability of an individual to solve IM-dominated tasks depends on cognitive mechanisms that operate in common across sensory modalities.

An Effect of Gaze Direction in Cocktail Party Listening

It is well established that gaze direction can influence auditory spatial perception, but the implications of this interaction for performance in complex listening tasks is unclear. In the current study, we investigated whether there is a measurable effect of gaze direction on speech intelligibility in a "cocktail party" listening situation. We presented sequences of digits from five loudspeakers positioned at 0°,  ±  15°, and ± 30° azimuth, and asked participants to repeat back the digits presented from a designated target loudspeaker. In different blocks of trials, the participant visually fixated on a cue presented at the target location or at a nontarget location. Eye position was tracked continuously to monitor compliance. Performance was best when fixation was on-target (vs. off-target) and the size of this effect depended on the specific configuration. This result demonstrates an influence of gaze direction in multitalker mixtures, even in the absence of visual speech information.

An ecological investigation of the capacity to follow simultaneous speech and preferential detection of ones’ own name

Many situations require focusing attention on one speaker, while monitoring the environment for potentially important information. Some have proposed that dividing attention among 2 speakers involves behavioral trade-offs, due to limited cognitive resources. However the severity of these trade-offs, particularly under ecologically-valid circumstances, is not well understood. We investigated the capacity to process simultaneous speech using a dual-task paradigm simulating task-demands and stimuli encountered in real-life. Participants listened to conversational narratives (Narrative Stream) and monitored a stream of announcements (Barista Stream), to detect when their order was called. We measured participants’ performance, neural activity, and skin conductance as they engaged in this dual-task. Participants achieved extremely high dual-task accuracy, with no apparent behavioral trade-offs. Moreover, robust neural and physiological responses were observed for target-stimuli in the Barista Stream, alongside significant neural speech-tracking of the Narrative Stream. These results suggest that humans have substantial capacity to process simultaneous speech and do not suffer from insufficient processing resources, at least for this highly ecological task-combination and level of perceptual load. Results also confirmed the ecological validity of the advantage for detecting ones’ own name at the behavioral, neural, and physiological level, highlighting the contribution of personal relevance when processing simultaneous speech.

Auditory motion as a cue for source segregation and selection in a "cocktail party" listening environment

Source motion was examined as a cue for segregating concurrent speech or noise sources. In two different headphone-based tasks-motion detection (MD) and speech-on-speech masking (SI)-one source among three was designated as the target only by imposing sinusoidal variation in azimuth during the stimulus presentation. For MD, the lstener was asked which of the three concurrent sources was in motion during the trial. For SI, the listener was asked to report the words spoken by the moving speech source. MD performance improved as the amplitude of the sinusoidal motion (i.e., displacement in azimuth) increased over the range of values tested (±5° to ±30°) for both modulated noise and speech targets, with better performance found for speech. SI performance also improved as the amplitude of target motion increased. Furthermore, SI performance improved as word position progressed throughout the sentence. Performance on the MD task was correlated with performance on SI task across individual subjects. For the SI conditions tested here, these findings are consistent with the proposition that listeners first detect the moving target source, then focus attention on the target location as the target sentence unfolds.

Sound Source Selection Based on Head Movements in Natural Group Conversation

To optimally improve signal-to-noise ratio in noisy environments, a hearing assistance device must correctly identify what is signal and what is noise. Many of the biosignal-based approaches to solving this question are themselves subject to noise, but head angle is an overt behavior that may be possible to capture in practical devices in the real world. Previous orientation studies have demonstrated that head angle is systematically related to listening target; our study aimed to examine whether this relationship is sufficiently reliable to be used in group conversations where participants may be seated in different layouts and the listener is free to turn their body as well as their head. In addition to this simple method, we developed a source-selection algorithm based on a hidden Markov model (HMM) trained on listeners’ head movement. The performance of this model and the simple head-steering method was evaluated using publicly available behavioral data. Head angle during group conversation was predictive of active talker, exhibiting an undershoot with a slope consistent with that found in simple orientation studies, but the intercept of the linear relationship was different for different talker layouts, suggesting it would be problematic to rely exclusively on this information to predict the location of auditory attention. Provided the location of all target talkers is known, the HMM source selection model implemented here, however, showed significantly lower error in identifying listeners’ auditory attention than the linear head-steering method.

Potential of Augmented Reality Platforms to Improve Individual Hearing Aids and to Support More Ecologically Valid Research

An augmented reality (AR) platform combines several technologies in a system that can render individual “digital objects” that can be manipulated for a given purpose. In the audio domain, these may, for example, be generated by speaker separation, noise suppression, and signal enhancement. Access to the “digital objects” could be used to augment auditory objects that the user wants to hear better. Such AR platforms in conjunction with traditional hearing aids may contribute to closing the gap for people with hearing loss through multimodal sensor integration, leveraging extensive current artificial intelligence research, and machine-learning frameworks. This could take the form of an attention-driven signal enhancement and noise suppression platform, together with context awareness, which would improve the interpersonal communication experience in complex real-life situations. In that sense, an AR platform could serve as a frontend to current and future hearing solutions. The AR device would enhance the signals to be attended, but the hearing amplification would still be handled by hearing aids. In this article, suggestions are made about why AR platforms may offer ideal affordances to compensate for hearing loss, and how research-focused AR platforms could help toward better understanding of the role of hearing in everyday life.

Benefits of triple acoustic beamforming during speech-on-speech masking and sound localization for bilateral cochlear-implant users

Bilateral cochlear-implant (CI) users struggle to understand speech in noisy environments despite receiving some spatial-hearing benefits. One potential solution is to provide acoustic beamforming. A headphone-based experiment was conducted to compare speech understanding under natural CI listening conditions and for two non-adaptive beamformers, one single beam and one binaural, called "triple beam," which provides an improved signal-to-noise ratio (beamforming benefit) and usable spatial cues by reintroducing interaural level differences. Speech reception thresholds (SRTs) for speech-on-speech masking were measured with target speech presented in front and two maskers in co-located or narrow/wide separations. Numerosity judgments and sound-localization performance also were measured. Natural spatial cues, single-beam, and triple-beam conditions were compared. For CI listeners, there was a negligible change in SRTs when comparing co-located to separated maskers for natural listening conditions. In contrast, there were 4.9- and 16.9-dB improvements in SRTs for the beamformer and 3.5- and 12.3-dB improvements for triple beam (narrow and wide separations). Similar results were found for normal-hearing listeners presented with vocoded stimuli. Single beam improved speech-on-speech masking performance but yielded poor sound localization. Triple beam improved speech-on-speech masking performance, albeit less than the single beam, and sound localization. Thus, triple beam was the most versatile across multiple spatial-hearing domains.

Enhancing the perceptual segregation and localization of sound sources with a triple beamformer

A triple beamformer was developed to exploit the capabilities of the binaural auditory system. The goal was to enhance the perceptual segregation of spatially separated sound sources while preserving source localization. The triple beamformer comprised a variant of a standard single-channel beamformer that routes the primary beam output focused on the target source location to both ears. The triple beam algorithm adds two supplementary beams with the left-focused beam routed only to the left ear and the right-focused beam routed only to the right ear. The rationale for the approach is that the triple beam processing exploits sound source segregation in high informational masking (IM) conditions. Furthermore, the exaggerated interaural level differences produced by the triple beam are well-suited for categories of listeners (e.g., bilateral cochlear implant users) who receive limited benefit from interaural time differences. The performance with the triple beamformer was compared to normal binaural hearing (simulated using a Knowles Electronic Manikin for Auditory Research, G.R.A.S. Sound and Vibration, Holte, DK) and to that obtained from a single-channel beamformer. Source localization in azimuth and masked speech identification for multiple masker locations were measured for all three algorithms. Taking both localization and speech intelligibility into account, the triple beam algorithm was considered to be advantageous under high IM listening conditions.

Benefits of Beamforming With Local Spatial-Cue Preservation for Speech Localization and Segregation

A study was conducted to examine the benefits afforded by a signal-processing strategy that imposes the binaural cues present in a natural signal, calculated locally in time and frequency, on the output of a beamforming microphone array. Such a strategy has the potential to combine the signal-to-noise ratio advantage of beamforming with the perceptual benefit of spatialization to enhance performance in multitalker mixtures. Participants with normal hearing and with hearing loss were tested on both speech localization and speech-on-speech masking tasks. Performance for the spatialized beamformer was compared with that for three other conditions: a reference condition with no processing, a beamformer with no spatialization, and a hybrid beamformer that operates only in the high frequencies to preserve natural binaural cues in the low frequencies. Beamforming with full-bandwidth spatialization supported speech localization and produced better speech reception thresholds than the other conditions.

Assessing the benefit of acoustic beamforming for listeners with aphasia using modified psychoacoustic methods

Acoustic beamforming has been shown to improve identification of target speech in noisy listening environments for individuals with sensorineural hearing loss. This study examined whether beamforming would provide a similar benefit for individuals with aphasia (acquired neurological language impairment). The benefit of beamforming was examined for persons with aphasia (PWA) and age- and hearing-matched controls in both a speech masking condition and a speech-shaped, speech-modulated noise masking condition. Performance was measured when natural spatial cues were provided, as well as when the target speech level was enhanced via a single-channel beamformer. Because typical psychoacoustic methods may present substantial experimental confounds for PWA, clinically guided modifications of experimental procedures were determined individually for each PWA participant. Results indicated that the beamformer provided a significant overall benefit to listeners. On an individual level, both PWA and controls who exhibited poorer performance on the speech masking condition with spatial cues benefited from the beamformer, while those who achieved better performance with spatial cues did not. All participants benefited from the beamformer in the noise masking condition. The findings suggest that a spatially tuned hearing aid may be beneficial for older listeners with relatively mild hearing loss who have difficulty taking advantage of spatial cues.

The Dynamics of Attention Shifts Among Concurrent Speech in a Naturalistic Multi-speaker Virtual Environment

Focusing attention on one speaker on the background of other irrelevant speech can be a challenging feat. A longstanding question in attention research is whether and how frequently individuals shift their attention towards task-irrelevant speech, arguably leading to occasional detection of words in a so-called unattended message. However, this has been difficult to gauge empirically, particularly when participants attend to continuous natural speech, due to the lack of appropriate metrics for detecting shifts in internal attention. Here we introduce a new experimental platform for studying the dynamic deployment of attention among concurrent speakers, utilizing a unique combination of Virtual Reality (VR) and Eye-Tracking technology. We created a Virtual Café in which participants sit across from and attend to the narrative of a target speaker. We manipulated the number and location of distractor speakers by placing additional characters throughout the Virtual Café. By monitoring participant's eye-gaze dynamics, we studied the patterns of overt attention-shifts among concurrent speakers as well as the consequences of these shifts on speech comprehension. Our results reveal important individual differences in the gaze-pattern displayed during selective attention to speech. While some participants stayed fixated on a target speaker throughout the entire experiment, approximately 30% of participants frequently shifted their gaze toward distractor speakers or other locations in the environment, regardless of the severity of audiovisual distraction. Critically, preforming frequent gaze-shifts negatively impacted the comprehension of target speech, and participants made more mistakes when looking away from the target speaker. We also found that gaze-shifts occurred primarily during gaps in the acoustic input, suggesting that momentary reductions in acoustic masking prompt attention-shifts between competing speakers, in line with "glimpsing" theories of processing speech in noise. These results open a new window into understanding the dynamics of attention as they wax and wane over time, and the different listening patterns employed for dealing with the influx of sensory input in multisensory environments. Moreover, the novel approach developed here for tracking the locus of momentary attention in a naturalistic virtual-reality environment holds high promise for extending the study of human behavior and cognition and bridging the gap between the laboratory and real-life.

Evaluating the Performance of a Visually Guided Hearing Aid Using a Dynamic Auditory-Visual Word Congruence Task

OBJECTIVES

The "visually guided hearing aid" (VGHA), consisting of a beamforming microphone array steered by eye gaze, is an experimental device being tested for effectiveness in laboratory settings. Previous studies have found that beamforming without visual steering can provide significant benefits (relative to natural binaural listening) for speech identification in spatialized speech or noise maskers when sound sources are fixed in location. The aim of the present study was to evaluate the performance of the VGHA in listening conditions in which target speech could switch locations unpredictably, requiring visual steering of the beamforming. To address this aim, the present study tested an experimental simulation of the VGHA in a newly designed dynamic auditory-visual word congruence task.

DESIGN

Ten young normal-hearing (NH) and 11 young hearing-impaired (HI) adults participated. On each trial, three simultaneous spoken words were presented from three source positions (-30, 0, and 30 azimuth). An auditory-visual word congruence task was used in which participants indicated whether there was a match between the word printed on a screen at a location corresponding to the target source and the spoken target word presented acoustically from that location. Performance was compared for a natural binaural condition (stimuli presented using impulse responses measured on KEMAR), a simulated VGHA condition (BEAM), and a hybrid condition that combined lowpass-filtered KEMAR and highpass-filtered BEAM information (BEAMAR). In some blocks, the target remained fixed at one location across trials, and in other blocks, the target could transition in location between one trial and the next with a fixed but low probability.

RESULTS

Large individual variability in performance was observed. There were significant benefits for the hybrid BEAMAR condition relative to the KEMAR condition on average for both NH and HI groups when the targets were fixed. Although not apparent in the averaged data, some individuals showed BEAM benefits relative to KEMAR. Under dynamic conditions, BEAM and BEAMAR performance dropped significantly immediately following a target location transition. However, performance recovered by the second word in the sequence and was sustained until the next transition.

CONCLUSIONS

When performance was assessed using an auditory-visual word congruence task, the benefits of beamforming reported previously were generally preserved under dynamic conditions in which the target source could move unpredictably from one location to another (i.e., performance recovered rapidly following source transitions) while the observer steered the beamforming via eye gaze, for both young NH and young HI groups.

Improving Speech Intelligibility by Hearing Aid Eye-Gaze Steering: Conditions With Head Fixated in a Multitalker Environment

The behavior of a person during a conversation typically involves both auditory and visual attention. Visual attention implies that the person directs his or her eye gaze toward the sound target of interest, and hence, detection of the gaze may provide a steering signal for future hearing aids. The steering could utilize a beamformer or the selection of a specific audio stream from a set of remote microphones. Previous studies have shown that eye gaze can be measured through electrooculography (EOG). To explore the precision and real-time feasibility of the methodology, seven hearing-impaired persons were tested, seated with their head fixed in front of three targets positioned at −30°, 0°, and +30° azimuth. Each target presented speech from the Danish DAT material, which was available for direct input to the hearing aid using head-related transfer functions. Speech intelligibility was measured in three conditions: a reference condition without any steering, a condition where eye gaze was estimated from EOG measures to select the desired audio stream, and an ideal condition with steering based on an eye-tracking camera. The “EOG-steering” improved the sentence correct score compared with the “no-steering” condition, although the performance was still significantly lower than the ideal condition with the eye-tracking camera. In conclusion, eye-gaze steering increases speech intelligibility, although real-time EOG-steering still requires improvements of the signal processing before it is feasible for implementation in a hearing aid.

Introduction to the Research Symposium Forum

The purpose of this introduction is to provide an overview of the articles contained within this research forum of JSLHR. Each of these articles is based upon presentations from the 2016 ASHA Research Symposium.

Examination of a hybrid beamformer that preserves auditory spatial cues

A hearing-aid strategy that combines a beamforming microphone array in the high frequencies with natural binaural signals in the low frequencies was examined. This strategy attempts to balance the benefits of beamforming (improved signal-to-noise ratio) with the benefits of binaural listening (spatial awareness and location-based segregation). The crossover frequency was varied from 200 to 1200 Hz, and performance was compared to full-spectrum binaural and beamformer conditions. Speech intelligibility in the presence of noise or competing speech was measured in listeners with and without hearing loss. Results showed that the optimal crossover frequency depended on the listener and the nature of the interference.

The Benefit of a Visually Guided Beamformer in a Dynamic Speech Task

The aim of this study was to evaluate the performance of a visually guided hearing aid (VGHA) under conditions designed to capture some aspects of "real-world" communication settings. The VGHA uses eye gaze to steer the acoustic look direction of a highly directional beamforming microphone array. Although the VGHA has been shown to enhance speech intelligibility for fixed-location, frontal targets, it is currently not known whether these benefits persist in the face of frequent changes in location of the target talker that are typical of conversational turn-taking. Participants were 14 young adults, 7 with normal hearing and 7 with bilateral sensorineural hearing impairment. Target stimuli were sequences of 12 question-answer pairs that were embedded in a mixture of competing conversations. The participant's task was to respond via a key press after each answer indicating whether it was correct or not. Spatialization of the stimuli and microphone array processing were done offline using recorded impulse responses, before presentation over headphones. The look direction of the array was steered according to the eye movements of the participant as they followed a visual cue presented on a widescreen monitor. Performance was compared for a "dynamic" condition in which the target stimulus moved between three locations, and a "fixed" condition with a single target location. The benefits of the VGHA over natural binaural listening observed in the fixed condition were reduced in the dynamic condition, largely because visual fixation was less accurate.