speech recognition in noise
Recently Published Documents


TOTAL DOCUMENTS

230
(FIVE YEARS 52)

H-INDEX

29
(FIVE YEARS 2)

2021 ◽  
Vol 4 ◽  
Author(s):  
Alireza Goudarzi ◽  
Gemma Moya-Galé

The sophistication of artificial intelligence (AI) technologies has significantly advanced in the past decade. However, the observed unpredictability and variability of AI behavior in noisy signals is still underexplored and represents a challenge when trying to generalize AI behavior to real-life environments, especially for people with a speech disorder, who already experience reduced speech intelligibility. In the context of developing assistive technology for people with Parkinson's disease using automatic speech recognition (ASR), this pilot study reports on the performance of Google Cloud speech-to-text technology with dysarthric and healthy speech in the presence of multi-talker babble noise at different intensity levels. Despite sensitivities and shortcomings, it is possible to control the performance of these systems with current tools in order to measure speech intelligibility in real-life conditions.


Author(s):  
Gertjan Dingemanse ◽  
André Goedegebure

Purpose: This study aimed to evaluate the effect of speech recognition performance, working memory capacity (WMC), and a noise reduction algorithm (NRA) on listening effort as measured with pupillometry in cochlear implant (CI) users while listening to speech in noise. Method: Speech recognition and pupil responses (peak dilation, peak latency, and release of dilation) were measured during a speech recognition task at three speech-to-noise ratios (SNRs) with an NRA in both on and off conditions. WMC was measured with a reading span task. Twenty experienced CI users participated in this study. Results: With increasing SNR and speech recognition performance, (a) the peak pupil dilation decreased by only a small amount, (b) the peak latency decreased, and (c) the release of dilation after the sentences increased. The NRA had no effect on speech recognition in noise or on the peak or latency values of the pupil response but caused less release of dilation after the end of the sentences. A lower reading span score was associated with higher peak pupil dilation but was not associated with peak latency, release of dilation, or speech recognition in noise. Conclusions: In CI users, speech perception is effortful, even at higher speech recognition scores and high SNRs, indicating that CI users are in a chronic state of increased effort in communication situations. The application of a clinically used NRA did not improve speech perception, nor did it reduce listening effort. Participants with a relatively low WMC exerted relatively more listening effort but did not have better speech reception thresholds in noise.


Author(s):  
Julie Beadle ◽  
Jeesun Kim ◽  
Chris Davis

Purpose: Listeners understand significantly more speech in noise when the talker's face can be seen (visual speech) in comparison to an auditory-only baseline (a visual speech benefit). This study investigated whether the visual speech benefit is reduced when the correspondence between auditory and visual speech is uncertain and whether any reduction is affected by listener age (older vs. younger) and how severe the auditory signal is masked. Method: Older and younger adults completed a speech recognition in noise task that included an auditory-only condition and four auditory–visual (AV) conditions in which one, two, four, or six silent talking face videos were presented. One face always matched the auditory signal; the other face(s) did not. Auditory speech was presented in noise at −6 and −1 dB signal-to-noise ratio (SNR). Results: When the SNR was −6 dB, for both age groups, the standard-sized visual speech benefit reduced as more talking faces were presented. When the SNR was −1 dB, younger adults received the standard-sized visual speech benefit even when two talking faces were presented, whereas older adults did not. Conclusions: The size of the visual speech benefit obtained by older adults was always smaller when AV correspondence was uncertain; this was not the case for younger adults. Difficulty establishing AV correspondence may be a factor that limits older adults' speech recognition in noisy AV environments. Supplemental Material https://doi.org/10.23641/asha.16879549


2021 ◽  
Vol Publish Ahead of Print ◽  
Author(s):  
Lisette M. van Leeuwen ◽  
Thadé Goderie ◽  
Marieke F. van Wier ◽  
Birgit I. Lissenberg-Witte ◽  
Ulrike Lemke ◽  
...  

Author(s):  
Bruna S. Mussoi

Purpose Music training has been proposed as a possible tool for auditory training in older adults, as it may improve both auditory and cognitive skills. However, the evidence to support such benefits is mixed. The goal of this study was to determine the differential effects of lifelong musical training and working memory on speech recognition in noise, in older adults. Method A total of 31 musicians and nonmusicians aged 65–78 years took part in this cross-sectional study. Participants had a normal pure-tone average, with most having high-frequency hearing loss. Working memory (memory capacity) was assessed with the backward Digit Span test, and speech recognition in noise was assessed with three clinical tests (Quick Speech in Noise, Hearing in Noise Test, and Revised Speech Perception in Noise). Results Findings from this sample of older adults indicate that neither music training nor working memory was associated with differences on the speech recognition in noise measures used in this study. Similarly, duration of music training was not associated with speech-in-noise recognition. Conclusions Results from this study do not support the hypothesis that lifelong music training benefits speech recognition in noise. Similarly, an effect of working memory (memory capacity) was not apparent. While these findings may be related to the relatively small sample size, results across previous studies that investigated these effects have also been mixed. Prospective randomized music training studies may be able to better control for variability in outcomes associated with pre-existing and music training factors, as well as to examine the differential impact of music training and working memory for speech-in-noise recognition in older adults.


Author(s):  
Victoria Stenbäck ◽  
Erik Marsja ◽  
Mathias Hällgren ◽  
Björn Lyxell ◽  
Birgitta Larsby

Purpose The study aimed to investigate the relationship between speech recognition in noise, age, hearing ability, self-rated listening effort, inhibitory control (measured with the Swedish Hayling task), and working memory capacity (WMC; measured with the Reading Span test). Two different speech materials were used: the Hagerman test with low semantic context and Hearing in Noise Test sentences with high semantic context, masked with either energetic or informational maskers. Method A mixed design was used. Twenty-four young normally hearing ( M age = 25.6 years) and 24 older, for their age, normally hearing individuals ( M age = 60.6 years) participated in the study. Speech recognition in noise in both speech materials and self-rated effort in all four background maskers were correlated with inhibitory control and WMC. A linear mixed-effects model was set up to assess differences between the two different speech materials, the four different maskers used in the study, and if age and hearing ability affected performance in the speech materials or the various background noises. Results Results showed that high WMC was related to lower scores of self-rated listening effort for informational maskers, as well as better performance in speech recognition in noise when informational maskers were used. The linear mixed-effects model revealed differences in performance between the low-context and the high-context speech materials, and the various maskers used. Lastly, inhibitory control had some impact on performance in the low-context speech material when masked with an informational masker. Conclusion Different background noises, especially informational maskers, affect speech recognition and self-rated listening effort differently depending on age, hearing ability, and individual variation in WMC and inhibitory control.


2021 ◽  
Vol 10 (19) ◽  
pp. 4305
Author(s):  
Farnaz Matin ◽  
Eralp-Niyazi Artukarslan ◽  
Angelika Illg ◽  
Anke Lesinski-Schiedat ◽  
Thomas Lenarz ◽  
...  

This retrospective study aimed to investigate the range of hearing levels in a cochlear implant (CI) elderly population receiving electric-acoustic-stimulation (EAS) or electric-stimulation (ES) alone. The investigation evaluates the degree of hearing preservation (HP) and the speech comprehension resulting from EAS or ES-only to identify audiometric factors that predict adequate EAS and ES use. We analyzed the pure tone audiometry and speech perception in quiet and noise preoperatively and 12-months after activation of 89 elderly adults (age of 65 years old or older), yielding in total 97 CIs. Thirty-two (33.1%) patients were potential EAS candidates preoperatively, of which 18 patients used EAS at the time of first fitting and the other 14 patients continued to use their residual hearing for EAS at 12-months. Post-treatment, patients with EAS system and ES-only users’ with longer electrodes showed better results in monosyllable word scores in quiet than ES-only users with shorter electrodes. A similar trend was revealed for the speech recognition in noise. Patients with an EAS system benefit from maintaining their natural residual hearing. Nevertheless, strict preoperative patient selection is warranted particularly in elderly patients, in whom the hearing thresholds for EAS indication differ slightly from that in younger adults.


Author(s):  
Soňa Šikolová ◽  
Milan Urík ◽  
Dagmar Hošnová ◽  
Vít Kruntorád ◽  
Michal Bartoš ◽  
...  

Abstract Purpose The study aimed to evaluate audiological benefits, quality of hearing and safety of two Bonebridge generation: BCI601 and BCI602 (MED-EL, Innsbruck, Austria) in children. Methods Twelve children were implanted: five BCI601 and seven BCI602 comprising of ten conductive hearing loss, and two single sided deaf SSD subjects. Audiological outcomes tested were sound field audiometry, functional gain, speech recognition threshold (SRT50), speech recognition in noise (SPRINT) and localisation abilities. Subjective measures were Speech, Spatial and Qualities of Hearing Scale (SSQ12). Results The mean FG with the BCI601 was 25.0 dB and with the BCI602 28.0 dB. The benefit in SRT50 was 23.2 dB and 33.8 dB, respectively. The mean benefit in SPRINT was 15% and 6.7% and the localisation ability improved from 33.3° to 16° and from 26.2° to 17.6°, respectively. The two SSD subjects reported a FG of 17 dB, a benefit in SRT50 of 22.5 and a benefit in SPRINT of 20%. Subjective outcomes improved significantly and even exceeded the values of their age-and sex matched normal hearing peers. One revision was reported: a retroauricular emphysema above the implant occurred 12 months post-OP, it was resolved operatively with the implant still being functional. Conclusion The pediatric cohort reports significant audiological benefit, even exceeding that of the age- and sex matched control. The combination of the high safety and audiological benefit makes the Bonebridge a comfortable and effective option in hearing rehabilitation in children.


Sign in / Sign up

Export Citation Format

Share Document