Talker Voice Research Articles

The structure of acoustic voice variation in bilingual speech.

When a bilingual switches languages, do they switch their voice? Using a conversational corpus of speech from early Cantonese-English bilinguals (n = 34), this paper examines the talker-specific acoustic signatures of bilingual voices. Following the psychoacoustic model of voice, 24 filter and source-based acoustic measurements are estimated. The analysis summarizes mean differences for these dimensions and identifies the underlying structure of each talker's voice across languages with principal component analyses. Canonical redundancy analyses demonstrate that while talkers vary in the degree to which they have the same voice across languages, all talkers show strong similarity with themselves, suggesting an individual's voice remains relatively constant across languages. Voice variability is sensitive to sample size, and we establish the required sample to settle on a consistent impression of one's voice. These results have implications for human and machine voice recognition for bilinguals and monolinguals and speak to the substance of voice prototypes.

Conceptualising acoustic and cognitive contributions to divided-attention listening within a data-limit versus resource-limit framework

An understanding of how listeners divide their attention between two simultaneous talkers requires modelling the interaction between acoustic factors (energetic masking) and cognitive processes (control of auditory attention). The impact of spatial separation between the two talkers on this interaction is unclear, since separation is likely to create both acoustic benefits (release from energetic masking) and cognitive costs (increased demands on spatial attentional control). To explore this question, we manipulated the degree of energetic masking (high vs. low) and spatial separation (collocated to dichotic) between two simultaneous talkers. When energetic masking was high (Experiment 1, unmanipulated talker voices), transcription performance improved monotonically from collocated to dichotic, owing to a gradual release from energetic masking. When energetic masking was low (Experiment 2, bandpass-filtered talker voices), the benefit of spatial separation disappeared; performance even worsened in the dichotic condition. Additionally, across both experiments, individual differences in working memory best predicted transcription performance in conditions where energetic masking was low. These results suggest that energetic masking is the dominant challenge during divided-attention listening, but that the contribution of cognitive control and working memory can be observed when energetic masking is reduced, at least in the context of the current paradigm. The findings are discussed in light of Norman and Bobrow’s (1975) concept of data-limited vs. resource-limited tasks, which we propose is a promising framework for reinterpreting existing results from speech-in-noise perception research.

Talker-specificity and token-specificity in recognition memory

Given any feasible amount of time, a talker would never be able to produce the same word twice in an identical manner. Yet recognition memory experiments have consistently used identical tokens to demonstrate that listeners recognize a word more quickly and accurately when it is repeated by the same talker than by a different talker. These talker-specificity effects have served as the foundation of decades of research in speech perception, but the use of identical tokens introduces a confound: Is it the talker or the physical stimulus that drives these effects? And consequently, to what extent do listeners encode the high-level acoustic characteristics of a talker's voice? We investigate the roles of token and talker repetition in two continuous recognition memory experiments. In Exp. 1, listeners heard the voice of one talker, with either Identical or Novel repeated tokens. In Exp. 2, listeners heard two demographically matched talkers, with same-voice repetitions being either Identical or Novel. Classic talker-specificity effects were replicated in both Identical and Novel tokens, but recognition of Identical tokens was in some cases stronger than recognition of Novel tokens. In addition, recognition memory varied across demographically matched talkers, suggesting stronger episodic encoding for one talker than for the other. We argue that novel tokens should serve as the default design for similar studies and that consideration of talker variation can advance our understanding of encoding and memory differences more broadly.

Short Implicit Voice Training Affects Listening Effort During a Voice Cue Sensitivity Task With Vocoder-Degraded Speech.

Understanding speech in real life can be challenging and effortful, such as in multiple-talker listening conditions. Fundamental frequency ( fo ) and vocal-tract length ( vtl ) voice cues can help listeners segregate between talkers, enhancing speech perception in adverse listening conditions. Previous research showed lower sensitivity to fo and vtl voice cues when speech signal was degraded, such as in cochlear implant hearing and vocoder-listening compared to normal hearing, likely contributing to difficulties in understanding speech in adverse listening. Nevertheless, when multiple talkers are present, familiarity with a talker's voice, via training or exposure, could provide a speech intelligibility benefit. In this study, the objective was to assess how an implicit short-term voice training could affect perceptual discrimination of voice cues ( fo+vtl ), measured in sensitivity and listening effort, with or without vocoder degradations. Voice training was provided via listening to a recording of a book segment for approximately 30 min, and answering text-related questions, to ensure engagement. Just-noticeable differences (JNDs) for fo+vtl were measured with an odd-one-out task implemented as a 3-alternative forced-choice adaptive paradigm, while simultaneously collecting pupil data. The reference voice either belonged to the trained voice or an untrained voice. Effects of voice training (trained and untrained voice), vocoding (non-vocoded and vocoded), and item variability (fixed or variable consonant-vowel triplets presented across three items) on voice cue sensitivity ( fo+vtl JNDs) and listening effort (pupillometry measurements) were analyzed. Results showed that voice training did not have a significant effect on voice cue discrimination. As expected, fo+vtl JNDs were significantly larger for vocoded conditions than for non-vocoded conditions and with variable item presentations than fixed item presentations. Generalized additive mixed models analysis of pupil dilation over the time course of stimulus presentation showed that pupil dilation was significantly larger during fo+vtl discrimination while listening to untrained voices compared to trained voices, but only for vocoder-degraded speech. Peak pupil dilation was significantly larger for vocoded conditions compared to non-vocoded conditions and variable items increased the pupil baseline relative to fixed items, which could suggest a higher anticipated task difficulty. In this study, even though short voice training did not lead to improved sensitivity to small fo+vtl voice cue differences at the discrimination threshold level, voice training still resulted in reduced listening effort for discrimination among vocoded voice cues.

Performance in an Audiovisual Selective Attention Task Using Speech-Like Stimuli Depends on the Talker Identities, But Not Temporal Coherence.

Audiovisual integration of speech can benefit the listener by not only improving comprehension of what a talker is saying but also helping a listener select a particular talker's voice from a mixture of sounds. Binding, an early integration of auditory and visual streams that helps an observer allocate attention to a combined audiovisual object, is likely involved in processing audiovisual speech. Although temporal coherence of stimulus features across sensory modalities has been implicated as an important cue for non-speech stimuli (Maddox et al., 2015), the specific cues that drive binding in speech are not fully understood due to the challenges of studying binding in natural stimuli. Here we used speech-like artificial stimuli that allowed us to isolate three potential contributors to binding: temporal coherence (are the face and the voice changing synchronously?), articulatory correspondence (do visual faces represent the correct phones?), and talker congruence (do the face and voice come from the same person?). In a trio of experiments, we examined the relative contributions of each of these cues. Normal hearing listeners performed a dual task in which they were instructed to respond to events in a target auditory stream while ignoring events in a distractor auditory stream (auditory discrimination) and detecting flashes in a visual stream (visual detection). We found that viewing the face of a talker who matched the attended voice (i.e., talker congruence) offered a performance benefit. We found no effect of temporal coherence on performance in this task, prompting an important recontextualization of previous findings.

Clearly, fame isn't everything: Talker familiarity does not augment talker adaptation.

Familiarity with a talker's voice provides numerous benefits to speech perception, including faster responses and improved intelligibility in quiet and in noise. Yet, it is unclear whether familiarity facilitates talker adaptation, or the processing benefit stemming from hearing speech from one talker compared to multiple different talkers. Here, listeners completed a speeded recognition task for words presented in either single-talker or multiple-talker blocks. Talkers were either famous (the last five Presidents of the United States of America) or non-famous (other male politicians of similar ages). Participants either received no information about the talkers before the word recognition task (Experiments 1 and 3) or heard the talkers and saw their names first (Experiment 2). As expected, responses were faster in the single-talker blocks than in the multiple-talker blocks. Famous voices elicited faster responses in Experiment 1, but familiarity effects were extinguished in Experiment 2, possibly by hearing all voices recently before the experiment. When talkers were counterbalanced across single-talker and mixed-talker blocks in Experiment 3, no familiarity effects were observed. Predictions of familiarity facilitating talker adaptation (smaller increase in response times across single- and multiple-talker blocks for famous voices) were not confirmed. Thus, talker familiarity might not augment adaptation to a consistent talker.

Interaction of bottom-up and top-down neural mechanisms in spatial multi-talker speech perception

How the human auditory cortex represents spatially separated simultaneous talkers and how talkers' locations and voices modulate the neural representations of attended and unattended speech are unclear. Here, we measured the neural responses from electrodes implanted in neurosurgical patients as they performed single-talker and multi-talker speech perception tasks. We found that spatial separation between talkers caused a preferential encoding of the contralateral speech in Heschl's gyrus (HG), planum temporale (PT), and superior temporal gyrus (STG). Location and spectrotemporal features were encoded in different aspects of the neural response. Specifically, the talker's location changed the mean response level, whereas the talker's spectrotemporal features altered the variation of response around response's baseline. These components were differentially modulated by the attended talker's voice or location, which improved the population decoding of attended speech features. Attentional modulation due to the talker's voice only appeared in the auditory areas with longer latencies, but attentional modulation due to location was present throughout. Our results show that spatial multi-talker speech perception relies upon a separable pre-attentive neural representation, which could be further tuned by top-down attention to the location and voice of the talker.

Influence of voice properties on vowel perception depends on speaker context.

Different speakers produce the same intended vowel with very different physical properties. Fundamental frequency (F0) and formant frequencies (FF), the two main parameters that discriminate between voices, also influence vowel perception. While it has been shown that listeners comprehend speech more accurately if they are familiar with a talker's voice, it is still unclear how such prior information is used when decoding the speech stream. In three online experiments, we examined the influence of speaker context via F0 and FF shifts on the perception of /o/-/u/ vowel contrasts. Participants perceived vowels from an /o/-/u/ continuum shifted toward /u/ when F0 was lowered or FF increased relative to the original speaker's voice and vice versa. This shift was reduced when the speakers were presented in a block-wise context compared to random order. Conversely, the original base voice was perceived to be shifted toward /u/ when presented in the context of a low F0 or high FF speaker, compared to a shift toward /o/ with high F0 or low FF speaker context. These findings demonstrate that that F0 and FF jointly influence vowel perception in speaker context.

Increased Connectivity among Sensory and Motor Regions during Visual and Audiovisual Speech Perception.

In everyday conversation, we usually process the talker's face as well as the sound of the talker's voice. Access to visual speech information is particularly useful when the auditory signal is degraded. Here, we used fMRI to monitor brain activity while adult humans (n = 60) were presented with visual-only, auditory-only, and audiovisual words. The audiovisual words were presented in quiet and in several signal-to-noise ratios. As expected, audiovisual speech perception recruited both auditory and visual cortex, with some evidence for increased recruitment of premotor cortex in some conditions (including in substantial background noise). We then investigated neural connectivity using psychophysiological interaction analysis with seed regions in both primary auditory cortex and primary visual cortex. Connectivity between auditory and visual cortices was stronger in audiovisual conditions than in unimodal conditions, including a wide network of regions in posterior temporal cortex and prefrontal cortex. In addition to whole-brain analyses, we also conducted a region-of-interest analysis on the left posterior superior temporal sulcus (pSTS), implicated in many previous studies of audiovisual speech perception. We found evidence for both activity and effective connectivity in pSTS for visual-only and audiovisual speech, although these were not significant in whole-brain analyses. Together, our results suggest a prominent role for cross-region synchronization in understanding both visual-only and audiovisual speech that complements activity in integrative brain regions like pSTS.SIGNIFICANCE STATEMENT In everyday conversation, we usually process the talker's face as well as the sound of the talker's voice. Access to visual speech information is particularly useful when the auditory signal is hard to understand (e.g., background noise). Prior work has suggested that specialized regions of the brain may play a critical role in integrating information from visual and auditory speech. Here, we show a complementary mechanism relying on synchronized brain activity among sensory and motor regions may also play a critical role. These findings encourage reconceptualizing audiovisual integration in the context of coordinated network activity.

Error in the Superior Temporal Gyrus? A Systematic Review and Activation Likelihood Estimation Meta-Analysis of Speech Production Studies.

Evidence for perceptual processing in models of speech production is often drawn from investigations in which the sound of a talker's voice is altered in real time to induce "errors." Methods of acoustic manipulation vary but are assumed to engage the same neural network and psychological processes. This paper aims to review fMRI and PET studies of altered auditory feedback and assess the strength of the evidence these studies provide for a speech error correction mechanism. Studies included were functional neuroimaging studies of speech production in neurotypical adult humans, using natural speech errors or one of three predefined speech manipulation techniques (frequency altered feedback, delayed auditory feedback, and masked auditory feedback). Seventeen studies met the inclusion criteria. In a systematic review, we evaluated whether each study (1) used an ecologically valid speech production task, (2) controlled for auditory activation caused by hearing the perturbation, (3) statistically controlled for multiple comparisons, and (4) measured behavioral compensation correlating with perturbation. None of the studies met all four criteria. We then conducted an activation likelihood estimation meta-analysis of brain coordinates from 16 studies that reported brain responses to manipulated over unmanipulated speech feedback, using the GingerALE toolbox. These foci clustered in bilateral superior temporal gyri, anterior to cortical fields typically linked to error correction. Within the limits of our analysis, we conclude that existing neuroimaging evidence is insufficient to determine whether error monitoring occurs in the posterior superior temporal gyrus regions proposed by models of speech production.

Adults Show Initial Advantages Over Children in Learning Difficult Nonnative Speech Sounds.

Purpose Children and early adolescents seem to have an advantage over adults in acquiring nonnative speech sounds, supported by evidence showing that earlier age of acquisition strongly predicts second language attainment. Although many factors influence children's ultimate success in language learning, it is unknown whether children rely on different, perhaps more efficient learning mechanisms than adults. Method The current study compared children (aged 10-16 years) and adults in their learning of a nonnative Hindi contrast. We tested the hypothesis that younger participants would show superior baseline discriminability or learning of the contrast, better memory for new sounds after a delay, or improved generalization to a new talker's voice. Measures of phonological and auditory skills were collected to determine whether individual variability in these skills predicts nonnative speech sound learning and whether these potential relationships differ between adults and children. Results Adults showed superior pretraining sensitivity to the contrast compared to children, and these pretraining discrimination scores predicted learning and retention. Even though adults seemed to have an initial advantage in learning, children improved after a period of off-line consolidation on the trained identification task and began to catch up to adults after an overnight delay. Additionally, perceptual skills that predicted speech sound learning differed between adults and children, suggesting they rely on different learning mechanisms. Conclusions These findings challenge the view that children are simply better speech sound learners than adults and suggest that their advantages may be due to different learning mechanisms or better retention of nonnative contrasts over the broader language learning trajectory. Supplemental Material https://doi.org/10.23641/asha.12735914.

Static and dynamic cocktail party listening in younger and older adults

Verbal communication often takes place in situations with several simultaneous speakers (“cocktail party listening”). These situations can be static (only one listening target) or dynamic (with alternating targets). In particular, dynamic cocktail party listening is believed to generate extra cognitive load and appears to be particularly demanding for older listeners.Two groups of younger and older listeners with good hearing and normal cognition participated in the present study. Three different, spatially separated talker voices uttering matrix sentences were presented to each listener with varying types and probabilities of target switches. Moreover, several neuropsychological tests were conducted to investigate general cognitive characteristics that may be important for speech understanding in these situations.In a static condition with a priori knowledge of the target talker, both age groups revealed very high speech recognition performance. In comparison, dynamic conditions caused extra costs associated with the need to monitor different potential sound sources and to refocus attention when the target changed. The amount of costs depended on the probability and type of target talker alterations. Again, no significant age-group differences were found. No significant associations of cognitive characteristics and costs could be shown. However, a more fine-grained analysis based on the calculation of general and specific switch costs showed different mechanisms in older and younger listeners.This study confirms that dynamic cocktail party listening is associated with costs that depend on the type and probability of target switches. It extends previous research by showing that the effects of switching type and probability are similar for younger and older listeners with good hearing and good cognitive abilities. It further shows that, despite comparable costs of dynamic listening, mechanisms are different for the two age groups, as switching auditory attention may be preserved with aging, but monitoring different sound sources appears to be more difficult for older adults.

Suprathreshold Differences in Competing Speech Perception in Older Listeners With Normal and Impaired Hearing.

Purpose Age-related declines in auditory temporal processing and cognition make older listeners vulnerable to interference from competing speech. This vulnerability may be increased in older listeners with sensorineural hearing loss due to additional effects of spectral distortion and accelerated cognitive decline. The goal of this study was to uncover differences between older hearing-impaired (OHI) listeners and older normal-hearing (ONH) listeners in the perceptual encoding of competing speech signals. Method Age-matched groups of 10 OHI and 10 ONH listeners performed the coordinate response measure task with a synthetic female target talker and a male competing talker at a target-to-masker ratio of +3 dB. Individualized gain was provided to OHI listeners. Each listener completed 50 baseline and 800 "bubbles" trials in which randomly selected segments of the speech modulation power spectrum (MPS) were retained on each trial while the remainder was filtered out. Average performance was fixed at 50% correct by adapting the number of segments retained. Multinomial regression was used to estimate weights showing the regions of the MPS associated with performance (a "classification image" or CImg). Results The CImg weights were significantly different between the groups in two MPS regions: a region encoding the shared phonetic content of the two talkers and a region encoding the competing (male) talker's voice. The OHI listeners demonstrated poorer encoding of the phonetic content and increased vulnerability to interference from the competing talker. Individual differences in CImg weights explained over 75% of the variance in baseline performance in the OHI listeners, whereas differences in high-frequency pure-tone thresholds explained only 10%. Conclusion Suprathreshold deficits in the encoding of low- to mid-frequency (~5-10 Hz) temporal modulations-which may reflect poorer "dip listening"-and auditory grouping at a perceptual and/or cognitive level are responsible for the relatively poor performance of OHI versus ONH listeners on a different-gender competing speech task. Supplemental Material https://doi.org/10.23641/asha.12568472.

Talker Processing in Mandarin-Speaking Congenital Amusics.

Purpose The ability to recognize individuals from their vocalizations is an important trait of human beings. In the current study, we aimed to examine how congenital amusia, an inborn pitch-processing disorder, affects discrimination and identification of talkers' voices. Method Twenty Mandarin-speaking amusics and 20 controls were tested on talker discrimination and identification in four types of contexts that varied in the degree of language familiarity: Mandarin real words, Mandarin pseudowords, Arabic words, and reversed Mandarin speech. Results The language familiarity effect was more evident in the talker identification task than the discrimination task for both participant groups, and talker identification accuracy decreased as native phonological representations were removed from the stimuli. Importantly, amusics demonstrated degraded performance in both native speech conditions that contained phonological/linguistic information to facilitate talker identification and nonnative conditions where talker voice processing primarily relied on phonetics cues, including pitch. Moreover, the performance in talker processing can be predicted by the participants' musical ability and phonological memory capacity. Conclusions The results provided a first set of behavioral evidence that individuals with amusia are impaired in the ability of human voice identification. Meanwhile, it is found that amusia is not only a pitch disorder but is likely to affect the phonological processing of speech, in terms of using phonological information in native speech to analyze a talker's identity. The above findings expanded the understanding of the nature and scope of congenital amusia. Supplemental Material https://doi.org/10.23641/asha.12170379.

Talker variability in word recognition under cochlear implant simulation: Does talker gender matter?

Normal-hearing listeners are less accurate and slower to recognize words with trial-to-trial talker changes compared to a repeating talker. Cochlear implant (CI) users demonstrate poor discrimination of same-gender talkers and, to a lesser extent, different-gender talkers, which could affect word recognition. The effects of talker voice differences on word recognition were investigated using acoustic noise-vocoder simulations of CI hearing. Word recognition accuracy was lower for multiple female and male talkers, compared to multiple female talkers or a single talker. Results suggest that talker variability has a detrimental effect on word recognition accuracy under CI simulation, but only with different-gender talkers.

Word Identification With Temporally Interleaved Competing Sounds by Younger and Older Adult Listeners.

The purpose of this experiment was to contribute to our understanding of the nature of age-related changes in competing speech perception using a temporally interleaved task. Younger and older adults (n = 16/group) participated in this study. The target was a five-word sentence. The masker was one of the following: another five-word sentence; five brief samples of modulated noise; or five brief samples of environmental sounds. The stimuli were presented in a temporally interleaved manner, where the target and masker alternated in time, always beginning with the target. Word order was manipulated in the target (and in the masker during trials with interleaved words) to compare performance when the five words in each stream did versus did not create a syntactically correct sentence. Talker voice consistency also was examined by contrasting performance when each word in the target was spoken by the same talker or by different talkers; a similar manipulation was used for the masker when it consisted of words. Participants were instructed to repeat back the target words and ignore the intervening words or sounds. Participants also completed a subset of tests from the NIH Cognitive Toolbox. Performance on this interleaved task was significantly associated with listener age and with a metric of cognitive flexibility, but it was not related to the degree of high-frequency hearing loss. Younger adults' performance on this task was better than that of older adults, especially for words located toward the end of the sentence. Both groups of participants were able to take advantage of correct word order in the target, and both were negatively affected, to a modest extent, when the masker words were in correct syntactic order. The two groups did not differ in how phonetic similarity between target and masker words influenced performance, and interleaved environmental sounds or noise had only a minimal effect for all listeners. The most robust difference between listener groups was found for the use of voice consistency: older adults, as compared with younger adults, were less able to take advantage of a consistent target talker within a trial. Younger adults outperformed older adults when masker words were interleaved with target words. Results suggest that this difference was unlikely to be related to energetic masking and/or peripheral hearing loss. Rather, age-related changes in cognitive flexibility and problems encoding voice information appeared to underlie group differences. These results support the contention that, in real-life competing speech situations that produce both energetic and informational masking, older adults' problems are due to both peripheral and nonperipheral changes.

A Laboratory Study of the McGurk Effect in 324 Monozygotic and Dizygotic Twins.

Multisensory integration of information from the talker’s voice and the talker’s mouth facilitates human speech perception. A popular assay of audiovisual integration is the McGurk effect, an illusion in which incongruent visual speech information categorically changes the percept of auditory speech. There is substantial interindividual variability in susceptibility to the McGurk effect. To better understand possible sources of this variability, we examined the McGurk effect in 324 native Mandarin speakers, consisting of 73 monozygotic (MZ) and 89 dizygotic (DZ) twin pairs. When tested with 9 different McGurk stimuli, some participants never perceived the illusion and others always perceived it. Within participants, perception was similar across time (r = 0.55 at a 2-year retest in 150 participants) suggesting that McGurk susceptibility reflects a stable trait rather than short-term perceptual fluctuations. To examine the effects of shared genetics and prenatal environment, we compared McGurk susceptibility between MZ and DZ twins. Both twin types had significantly greater correlation than unrelated pairs (r = 0.28 for MZ twins and r = 0.21 for DZ twins) suggesting that the genes and environmental factors shared by twins contribute to individual differences in multisensory speech perception. Conversely, the existence of substantial differences within twin pairs (even MZ co-twins) and the overall low percentage of explained variance (5.5%) argues against a deterministic view of individual differences in multisensory integration.

The visual speech head start improves perception and reduces superior temporal cortex responses to auditory speech.

Visual information about speech content from the talker's mouth is often available before auditory information from the talker's voice. Here we examined perceptual and neural responses to words with and without this visual head start. For both types of words, perception was enhanced by viewing the talker's face, but the enhancement was significantly greater for words with a head start. Neural responses were measured from electrodes implanted over auditory association cortex in the posterior superior temporal gyrus (pSTG) of epileptic patients. The presence of visual speech suppressed responses to auditory speech, more so for words with a visual head start. We suggest that the head start inhibits representations of incompatible auditory phonemes, increasing perceptual accuracy and decreasing total neural responses. Together with previous work showing visual cortex modulation (Ozker et al., 2018b) these results from pSTG demonstrate that multisensory interactions are a powerful modulator of activity throughout the speech perception network.

Masculine toughness and the categorical perception of onset sibilant clusters.

Evidence from production studies demonstrate that /s/-retraction, a sound change in American English where /s/ approaches [ʃ], is progressing in apparent time and highly context-dependent, occurring overwhelmingly in /str/ clusters. This study reports findings from a phoneme categorization task, suggesting that listeners perceive sibilants in /str/ clusters less categorically than in /spr/ and /skr/ clusters and are becoming less categorical in apparent time for all clusters. Indicators of masculine toughness, including the model talker's voice and face and the listener's endorsement of masculine stereotypes, additionally contribute to a less categorical perception of sibilants in these environments.

Language Ability and the Familiar Talker Advantage: Generalizing to Unfamiliar Talkers Is What Matters.

Purpose Previous studies with children and adults have demonstrated a familiar talker advantage-better word recognition for familiar talkers. The goal of the current study was to test whether this phenomenon is modulated by a child's language ability. Method Sixty children with a range of language ability were trained to learn the voices of 3 foreign-accented, German-English bilingual talkers and received feedback about their performance. Both before and after this talker voice training, children completed a spoken word recognition task in which they heard consonant-vowel-consonant words mixed with noise that were spoken by the 3 familiarized talkers and by 3 unfamiliar German-English bilinguals. Results Two findings emerged from this study: First, children with both higher and lower language ability performed similarly on the familiarized talkers. Second, children with higher language scores performed similarly on both the familiarized and unfamiliar talkers, whereas children with lower language scores performed worse on the unfamiliar talkers compared to familiar talkers, suggesting an inability to generalize to novel, unfamiliar talkers who spoke with a similar accent. Discussion Together, these findings indicate that children with higher language scores are able to generalize knowledge about foreign-accented talkers to help spoken word recognition for novel talkers with the same accent. In contrast, children with lower language skills did not exhibit the same magnitude of generalization. This lack of generalization to similar talkers may mean that children with lower language skills are at a disadvantage in spoken language tasks because they are unable to process speech as well when listening to unfamiliar talkers.