Search PubMed⌕ Search
Home/Archive/PubMed

PubMed · 42678253

The voice clone intelligibility benefit in noise in middle-aged listeners.

Patti Adank·Han Wang
View original source ↗Export metadata ↓

Abstract

Research with younger adults showed that cloned voices are more intelligible than human voices in noise, with a benefit of 13.4%. This study tested whether this benefit extends to 40 middle-aged listeners (45-65 years), as this population may show emerging difficulties with speech-in-noise. Participants recognised sentences by ten human voices and ten voice clones in four noise levels. Cloned voices were 11.8% more intelligible, with benefits enhanced at the two most severe noise levels (15.9% at -6 dB and 17.5% at -3 dB), suggesting cloned speech enhanced perception in middle-aged listeners, potentially by reducing listening effort and compensating for emerging age-related auditory-cognitive decline.

Explore related subjects

HumansMiddle AgedNoiseSpeech PerceptionFemaleAgedSpeech IntelligibilityMaleVoiceJournal Article

Keep this discovery

BibTeXRIS

Patti Adank, Han Wang. 2026-09-01. The voice clone intelligibility benefit in noise in middle-aged listeners.. https://doi.org/10.1121/10.0046178

Cite the original work for its findings. Save a collection to share your selection of sources.

Discover connections

Journal Article resourceAged subjectFemale subjectHumans subjectJournal Article subjectMale subjectMiddle Aged subject

Connections use source metadata and explicit phrase matches, not verified experimental comparisons.

KEEP EXPLORING

Related citations

PubMed2026

Beyond Photometric Consistency: Addressing Loss Insensitivity to Depth Noise in Endoscopic Estimation via Error Calibration.

Self-supervised monocular depth estimation in endoscopy is fundamentally constrained by the ill-posed nature of photometric supervision. In this work, we identify a critical yet overlooked cause of this ambiguity: the inherent insensitivity of photometric loss to depth noise. To overcome this intrinsic limitation, we propose Depth Error Calibration Learning (DECL), a two-stage framework that suppresses prediction variance and mitigates residual errors in self-supervised depth estimation. In Stage I (Variance Reduction), a cyclic depth generation strategy produces multiple depth hypotheses for the input image. The per-pixel empirical variance is quantified and integrated into a dedicated variance loss term, which penalizes inconsistent predictions and encourages the network to generate more stable and reliable depth estimates. In Stage II (Bias Calibration), an image-conditioned diffusion model refines the Stage-I depth prior and mitigates structured residuals through iterative denoising, thereby improving geometric accuracy and global consistency. Extensive experiments on three public endoscopic datasets demonstrate that DECL achieves consistent improvements over representative self-supervised monocular depth estimation methods under the evaluated protocols. Moreover, ablation studies on two representative backbones indicate that DECL is not restricted to a single network implementation, while broader validation on additional backbone families remains necessary. The source code is publicly available at https://github.com/DavidLuBit/EndoDenoising.

Journal Article↗
PubMed2026

AI-driven snapshot hyperspectral imaging for on-line sorting systems in food industry: From real-time sensing to intelligent decision-making.

High-throughput food sorting requires rapid, non-destructive detection of external defects, foreign materials, and internal quality attributes in heterogeneous food matrices. Conventional scanning hyperspectral imaging may suffer from motion-induced spatial-spectral mismatches, whereas snapshot hyperspectral imaging (S-HSI) captures spectral images within a single integration time. However, its advantage is limited by trade-offs in resolution, signal-to-noise ratio (SNR), reconstruction uncertainty, and calibration stability, which are further amplified by variable tissue structure, surface reflection, moisture, and fat distribution in foods. This review critically examines artificial intelligence (AI)-driven S-HSI for on-line food sorting within a sensing-representation-decision-execution framework. Compact architectures are compared according to their physical constraints, food-sorting suitability, and ability to support mapping between spectral responses and physicochemical quality attributes. AI strategies are reviewed for spectral reconstruction, image restoration, spatial-spectral representation, band selection, uncertainty-aware decision-making, and edge implementation. AI can partially compensate for snapshot-specific limitations, but current evidence remains largely limited to laboratory or prototype studies. Future work should link system performance to food safety and quality outcomes by reporting throughput, decision latency, calibration drift, missed-detection risk, false-rejection cost, and closed-loop sorting success.

Hyperspectral Imaging↗
PubMed2026

Transient acoustic stimulation induces time-dependent synaptic remodeling and enhancement of auditory nerve output after threshold recovery.

BACKGROUND: Acoustic stress can alter cochlear function even in the absence of permanent threshold elevation; however, synaptic consequences of transient acoustic stimulation remain incompletely understood. OBJECTIVE: This study aimed to investigate whether transient acoustic stimulation induces changes in the auditory nerve output and cochlear ribbon synapse morphology following hearing threshold recovery. METHODS: Young adult CBA/CaJ mice were exposed to band-limited acoustic stimulation (45-2,000 Hz, 95 dB SPL, 2 h). Auditory brainstem responses (ABRs), hair cell and spiral ganglion neuron survival, and synaptic morphology were evaluated before exposure and up to 2 weeks post-exposure. RESULTS: ABR thresholds were transiently elevated immediately after exposure but largely recovered by 1 day post-exposure. In contrast, ABR wave I amplitudes significantly increased after threshold recovery across multiple test frequencies. Ribbon-associated puncta in both inner and outer hair cell regions exhibited biphasic temporal changes, with an initial decrease immediately after exposure followed by an increase at 1 day post-exposure. The ribbon-associated punctal area also increased after exposure and remained elevated at later post-exposure time points. No significant loss of hair cells or spiral ganglion neurons was observed. Exploratory genomic analysis suggested enrichment of pathways related to metabolic defense and cellular stress responses. CONCLUSIONS: Transient acoustic stimulation induces time-dependent synaptic remodeling and enhancement of peripheral auditory nerve output without overt cellular degeneration. These findings support a model in which early cochlear responses to acoustic perturbation include adaptive synaptic plasticity and gain regulation, extending current concepts of noise-induced cochlear change beyond irreversible synaptic loss.

Animals↗