Search

Child-directed speech (CDS) is a particular type of speech that adults use when addressing young children. Its properties also change as a function of extralinguistic factors, such as age of the child being addressed. Access to large amounts of representative and varied CDS would be useful for child language research, as this would enable controlled computational modeling experiments of infant language acquisition with realistic input in terms of quality and quantity. In this study, we describe an approach to model age-dependent linguistic properties of CDS using a language model (LM) trained on CDS transcripts and ages of the recipient children, as obtained from North American English corpora of the CHILDES database. The created LM can then be used to stochastically generate synthetic CDS transcripts in an age-appropriate manner, thereby scaling beyond the original datasets in size. We compare characteristics of the generated CDS against the real speech addressed at children of different ages, showing that the LM manages to capture age-dependent changes in CDS, except for a slight difference in the effective vocabulary size. As a side product, we also provide a systematic characterization of age-dependent linguistic properties of CDS in CHILDES, illustrating how all measured aspects of the CDS change with children's age.

Computational models of child language development can help us understand the cognitive underpinnings of the language learning process, which occurs along several linguistic levels at once (e.g., prosodic and phonological). However, in light of the replication crisis, modelers face the challenge of selecting representative and consolidated infant data. Thus, it is desirable to have evaluation methodologies that could account for robust empirical reference data, across multiple infant capabilities. Moreover, there is a need for practices that can compare developmental trajectories of infants to those of models as a function of language experience and development. The present study aims to take concrete steps to address these needs by introducing the concept of comparing models with large-scale cumulative empirical data from infants, as quantified by meta-analyses conducted across a large number of individual behavioral studies. We formalize the connection between measurable model and human behavior, and then present a conceptual framework for meta-analytic evaluation of computational models. We exemplify the meta-analytic model evaluation approach with two modeling experiments on infant-directed speech preference and native/non-native vowel discrimination.

Modelling of early language acquisition aims to understand how infants bootstrap their language skills. The modelling encompasses properties of the input data used for training the models, the cognitive hypotheses and their algorithmic implementations being tested, and the evaluation methodologies to compare models to human data. Recent developments have enabled the use of more naturalistic training data for computational models. This also motivates development of more naturalistic tests of model behaviour. A crucial step towards such an aim is to develop representative speech datasets consisting of speech heard by infants in their natural environments. However, a major drawback of such recordings is that they are typically noisy, and it is currently unclear how the sound quality could affect analyses and modelling experiments conducted on such data. In this paper, we explore this aspect for the case of infant-directed speech (IDS) and adult-directed speech (ADS) analysis. First, we manually and automatically annotated audio quality of utterances extracted from two corpora of child-centred long-form recordings (in English and French). We then compared acoustic features of IDS and ADS in an in-lab dataset and across different audio quality subsets of naturalistic data. Finally, we assessed how the audio quality and recording environment may change the conclusions of a modelling analysis using a recent self-supervised learning model. Our results show that the use of modest and high audio quality naturalistic speech data result in largely similar conclusions on IDS and ADS in terms of acoustic analyses and modelling experiments. We also found that an automatic sound quality assessment tool can be used to screen out useful parts of long-form recordings for a closer analysis with comparable results to that of manual quality annotation.

When assessing children in laboratory experiments, the measured responses also contain task-irrelevant participant-level variability (“noise”) and measurement noise. Since experimental data are used to make inferences of development of cognitive capabilities with age, it is important to know if reliability of the used measurements depends on child age. Any systematic age-dependent changes in reliability could result in misleading developmental trajectories, as lower reliability will necessarily result in smaller effect sizes. This paper examines age-dependency of task-independent measurement variability in early childhood (3–40 months) by analyzing two large-scale datasets of participant-level experimental responses: the ManyBabies infant-directed speech preference (MB-IDS) dataset and a saccadic reaction time (SRT) dataset collected from rural South Africa. Analysis of participant- and study-level data reveals that MB-IDS shows comparable reliability across the included age range. In contrast, SRTs reflect systematically increasing measurement consistency with increasing age. Potential reasons and implications of this divergence are briefly discussed.

Previous computational models of early language acquisition have shown how linguistic structure of speech can be acquired using auditory or audiovisual learning mechanisms. However, real infants have sustained access to both uni- and multimodal sensory experiences. Therefore, it is of interest how the uni- and multimodal learning mechanisms could operate in concert, and how their interplay might affect the acquisition dynamics of different linguistic representations. This paper explores these questions with a computational model capable of simultaneous auditory and audiovisual learning from speech and images. We study how the model’s latent representations reflect phonemic, lexical, and semantic knowledge as a function of language experience. We also test how the findings vary with differential emphasis on the two learning mechanisms. As a result, we find phonemic learning always starting to emerge before lexical learning, followed by semantics. However, there is also notable overlap in their development. The same pattern emerges irrespectively of the emphasis on auditory or audiovisual learning. The result illustrates how the acquisition dynamics of linguistic representations are decoupled from the primary learning objectives (mechanisms) of the learner, and how the emergence of phonemes and words can be facilitated by both auditory and audiovisual learning in a synergetic manner.

The exaggerated intonation and special rhythmic properties ofinfant-directed speech (IDS) have been hypothesized to attractinfant’s attention to the speech stream. However, studiesinvestigating IDS in the context of models of attention arefew. A number of such models suggest that surprising ornovel perceptual inputs attract attention, where novelty can beoperationalized as the statistical predictability of the stimulusin a context. Since prosodic patterns such as F0 contours areaccessible to young infants who are also adept statisticallearners, the present paper investigates a hypothesis that pitchcontours in IDS are less predictable than those in adult-directed speech (ADS), thereby efficiently tapping into thebasic attentional mechanisms of the listeners. Results fromanalyses with naturalistic IDS and ADS speech show that IDShas lower overall predictability of intonation acrossneighboring syllables even when the F0 contours in bothspeaking styles are normalized to the same frequency range.

Infants’ speech perception adapts to the phonemic categoriesof their native language, a process assumed to be driven bythe distributional properties of speech. This study investigateswhether deep neural networks (DNNs), the current state-of-the-art in distributional feature learning, are capable oflearning phoneme-like representations of speech in anunsupervised manner. We trained DNNs with unlabeled andlabeled speech and analyzed the activations of each layer withrespect to the phones in the input segments. The analysesreveal that the emergence of phonemic invariance in DNNs isdependent on the availability of phonemic labeling of theinput during the training. No increased phonemic selectivityof the hidden layers was observed in the purely unsupervisednetworks despite successful learning of low-dimensionalrepresentations for speech. This suggests that additionallearning constraints or more sophisticated models are neededto account for the emergence of phone-like categories indistributional learning operating on natural speech.

Recent work has proposed that prominence perception inspeech could be driven by predictability of prosodic patterns,connecting prominence perception to the concept of statisticallearning. In the present study, we tested the predictabilityhypothesis by conducting a listening test where subjects werefirst exposed to a 5-minute stream of sentences with a certainproportion of sentence-final words having either a falling orrising pitch trajectory. After the exposure stage, subjects wereasked to grade prominence in a set of novel sentences withsimilar pitch patterns. The results show that the subjects weresignificantly more likely to perceive words with low-probability pitch trajectories as prominent independently ofthe direction of the pitch change. This suggests that evenshort exposure to prosodic patterns with a certain statisticalstructure can induce changes in prominence perception,supporting the connection between prominence perceptionand attentional orientation towards low-probability events inan otherwise predictable context.

Numerous studies have examined the acoustic correlates of sentential stress and its underlying linguistic functionality. However, the mechanism that connects stress cues to the listener's attentional processing has remained unclear. Also, the learnability versus innateness of stress perception has not been widely discussed. In this work, we introduce a novel perspective to the study of sentential stress and put forward the hypothesis that perceived sentence stress in speech is related to the unpredictability of prosodic features, thereby capturing the attention of the listener. As predictability is based on the statistical structure of the speech input, the hypothesis also suggests that stress perception is a result of general statistical learning mechanisms. To study this idea, computational simulations are performed where temporal prosodic trajectories are modeled with an n-gram model. Probabilities of the feature trajectories are subsequently evaluated on a set of novel utterances and compared to human perception of stress. The results show that the low-probability regions of F0 and energy trajectories are strongly correlated with stress perception, giving support to the idea that attention and unpredictability of sensory stimulus are mutually connected.

Computational models of child language development can help us understand the cognitive underpinnings of the language learning process, which occurs along several linguistic levels at once (e.g., prosodic and phonological). However, in light of the replication crisis, modelers face the challenge of selecting representative and consolidated infant data. Thus, it is desirable to have evaluation methodologies that could account for robust empirical reference data, across multiple infant capabilities. Moreover, there is a need for practices that can compare developmental trajectories of infants to those of models as a function of language experience and development. The present study aims to take concrete steps to address these needs by introducing the concept of comparing models with large‐scale cumulative empirical data from infants, as quantified by meta‐analyses conducted across a large number of individual behavioral studies. We formalize the connection between measurable model and human behavior, and then present a conceptual framework for meta‐analytic evaluation of computational models. We exemplify the meta‐analytic model evaluation approach with two modeling experiments on infant‐directed speech preference and native/non‐native vowel discrimination. [ABSTRACT FROM AUTHOR]

Infant motility assessment using intelligent wearables is a promising new approach for assessment of infant neurophysiological development, and where efficient signal analysis plays a central role. This study investigates the use of different end-to-end neural network architectures for processing infant motility data from wearable sensors. We focus on the performance and computational burden of alternative sensor encoder and time series modeling modules and their combinations. In addition, we explore the benefits of data augmentation methods in ideal and nonideal recording conditions. The experiments are conducted using a dataset of multisensor movement recordings from 7-month-old infants, as captured by a recently proposed smart jumpsuit for infant motility assessment. Our results indicate that the choice of the encoder module has a major impact on classifier performance. For sensor encoders, the best performance was obtained with parallel two-dimensional convolutions for intrasensor channel fusion with shared weights for all sensors. The results also indicate that a relatively compact feature representation is obtainable for within-sensor feature extraction without a drastic loss to classifier performance. Comparison of time series models revealed that feedforward dilated convolutions with residual and skip connections outperformed all recurrent neural network (RNN)-based models in performance, training time, and training stability. The experiments also indicate that data augmentation improves model robustness in simulated packet loss or sensor dropout scenarios. In particular, signal- and sensor-dropout-based augmentation strategies provided considerable boosts to performance without negatively affecting the baseline performance. Overall, the results provide tangible suggestions on how to optimize end-to-end neural network training for multichannel movement sensor data. [ABSTRACT FROM AUTHOR]

Automatic word count estimation (WCE) from audio recordings can be used to quantify the amount of verbal communication in a recording environment. One key application of WCE is to measure language input heard by infants and toddlers in their natural environments, as captured by daylong recordings from microphones worn by the infants. Although WCE is nearly trivial for high-quality signals in high-resource languages, daylong recordings are substantially more challenging due to the unconstrained acoustic environments and the presence of near- and far-field speech. Moreover, many use cases of interest involve languages for which reliable ASR systems or even well-defined lexicons are not available. A good WCE system should also perform similarly for low- and high-resource languages in order to enable unbiased comparisons across different cultures and environments. Unfortunately, the current state-of-the-art solution, the LENA system, is based on proprietary software and has only been optimized for American English, limiting its applicability. In this paper, we build on existing work on WCE and present the steps we have taken towards a freely available system for WCE that can be adapted to different languages or dialects with a limited amount of orthographically transcribed speech data. Our system is based on language-independent syllabification of speech, followed by a language-dependent mapping from syllable counts (and a number of other acoustic features) to the corresponding word count estimates. We evaluate our system on samples from daylong infant recordings from six different corpora consisting of several languages and socioeconomic environments, all manually annotated with the same protocol to allow direct comparison. We compare a number of alternative techniques for the two key components in our system: speech activity detection and automatic syllabification of speech. As a result, we show that our system can reach relatively consistent WCE accuracy across multiple corpora and languages (with some limitations). In addition, the system outperforms LENA on three of the four corpora consisting of different varieties of English. We also demonstrate how an automatic neural network-based syllabifier, when trained on multiple languages, generalizes well to novel languages beyond the training data, outperforming two previously proposed unsupervised syllabifiers as a feature extractor for WCE. [ABSTRACT FROM AUTHOR]

Highlights • The role of spectral tilt in sentence prominence was investigated. • Estimators for tilt of glottal source spectrum and surface spectrum were compared. • Multidimensional representations for spectral tilt were also investigated. • Robustness of different tilt estimators was tested for noisy and coded speech. • Experiments were conducted on both French and Dutch speech data. Abstract Linguistic prominence in speech is known to correlate with the acoustic measures of energy, F0, and duration. In contrast, the role of spectral tilt in the realization of prominence has remained more inconsistent between previous empirical investigations. This may be partially due to the lack of a standard method for quantifying spectral tilt or due to difficulties in estimating the acoustical source of spectral tilt, the glottal flow, from continuous speech. These issues have rendered interpretations and comparisons between studies difficult. In addition, (i) little is known about the robustness of tilt estimators for prominence detection in the case when speech is not clean but corrupted, as in real life, by environmental noise or telephone transmission (i.e. degradation caused by bandpass filtering and quantization noise). Moreover, (ii) little attention has been paid to multidimensional representations of source spectrum that can potentially incorporate more information about the phonation style than purely scalar measures. In this work, we study spectral tilt in signaling prominence in spoken Dutch and French under different levels of additive noise, and for telephone-band coded speech, and compare several one-dimensional tilt measures that have been previously encountered in the literature as well as multidimensional tilt measures. We also compare spectral tilt measures with other standard acoustic correlates for prominence, namely, energy, F0, and duration. Our results provide further empirical support for the finding that tilt is a systematic correlate of prominence in Dutch, that the role is smaller in French, and that energy, F0, and duration appear still to be the most robust features for discriminating prominent and non-prominent words. In addition, our results show that there are notable differences between different tilt measures at different levels of noise, and that multidimensional representations for tilt improve class separability from the scalar measures. [ABSTRACT FROM AUTHOR]

The exaggerated intonation and special rhythmic properties of infant-directed speech (IDS) have been hypothesized to attract infants' attention to the speech stream. However, there has been little work actually connecting the properties of IDS to models of attentional processing or perceptual learning. A number of such attention models suggest that surprising or novel perceptual inputs attract attention, where novelty can be operationalized as the statistical (un)predictability of the stimulus in the given context. Since prosodic patterns such as F0 contours are accessible to young infants who are also known to be adept statistical learners, the present paper investigates a hypothesis that F0 contours in IDS are less predictable than those in adult-directed speech (ADS), given previous exposure to both speaking styles, thereby potentially tapping into basic attentional mechanisms of the listeners in a similar manner that relative probabilities of other linguistic patterns are known to modulate attentional processing in infants and adults. Computational modeling analyses with naturalistic IDS and ADS speech from matched speakers and contexts show that IDS intonation has lower overall temporal predictability even when the F0 contours of both speaking styles are normalized to have equal means and variances. A closer analysis reveals that there is a tendency of IDS intonation to be less predictable at the end of short utterances, whereas ADS exhibits more stable average predictability patterns across the full extent of the utterances. The difference between IDS and ADS persists even when the proportion of IDS and ADS exposure is varied substantially, simulating different relative amounts of IDS heard in different family and cultural environments. Exposure to IDS is also found to be more efficient for predicting ADS intonation contours in new utterances than exposure to the equal amount of ADS speech. This indicates that the more variable prosodic contours of IDS also generalize to ADS, and may therefore enhance prosodic learning in infancy. Overall, the study suggests that one reason behind infant preference for IDS could be its higher information value at the prosodic level, as measured by the amount of surprisal in the F0 contours. This provides the first formal link between the properties of IDS and the models of attentional processing and statistical learning in the brain. However, this finding does not rule out the possibility that other differences between the IDS and ADS also play a role. [ABSTRACT FROM AUTHOR]

Syllables are often considered to be central to infant and adult speech perception. Many theories and behavioral studies on early language acquisition are also based on syllable-level representations of spoken language. There is little clarity, however, on what sort of pre-linguistic "syllable" would actually be accessible to an infant with no phonological or lexical knowledge. Anchored by the notion that syllables are organized around particularly sonorous (audible) speech sounds, the present study investigates the feasibility of speech segmentation into syllable-like chunks without any a priori linguistic knowledge. We first operationalize sonority as a measurable property of the acoustic input, and then use sonority variation across time, or speech rhythm, as the basis for segmentation. The entire process from acoustic input to chunks of syllable-like acoustic segments is implemented as a computational model inspired by the oscillatory entrainment of the brain to speech rhythm. We analyze the output of the segmentation process in three different languages, showing that the sonority fluctuation in speech is highly informative of syllable and word boundaries in all three cases without any language-specific tuning of the model. These findings support the widely held assumption that syllable-like structure is accessible to infants even when they are only beginning to learn the properties of their native language. [ABSTRACT FROM AUTHOR]

When infants learn to pronounce speech sounds of their native language, they face the so-called correspondence problem – how to know which articulatory gestures lead to acoustic sounds that are recognized as native speech sounds by other speakers? Previous research suggests that infants might not learn to imitate their parents via autonomous babbling because direct evaluation of the acoustic similarity between the speech sounds of the two is not possible due to different spectral characteristics of the voices caused by differing vocal tract morphologies. We present a novel robust model of infant vowel imitation learning, following a hypothesis that an infant learns to match their productions to their caregiver's speech sounds when the caregiver imitates the infant's babbles. Adapting a cross-situational associative learning technique, evidently present in infant word learning, our simulated language learner can cope with ambiguity in caregiver's responses to babbling as well as with the imprecision of the articulatory gestures of the infant itself. Our fully online learning model also combines vocal exploration and imitative interaction into a single process. Learning performance is evaluated in experiments using Finnish adults as caregivers for a virtual infant, responding to the infant's babbles with lexical words and, after a learning stage, evaluating the quality of the vowels produced by the learner. After 1000 babble-response pairs, our virtual infant is seen to reach a satisfying vowel imitation accuracy of 70–80%. [ABSTRACT FROM AUTHOR]

Automatic detection of prominence in speech has attracted interest in recent years due to its multiple uses in spoken language applications. However, typical approaches require manual labeling of the data that is an expensive and time consuming process, also making the systems potentially specific to the language or speaking style in question. In this paper, we propose a novel unsupervised algorithm for the automatic detection of sentence prominence named 3PRO (Prominence from Prosodic Probabilities; “ three-pro ”) that is based on recent findings on human perception of prominence in speech. By combining syllable duration information to the level of surprisal observed in the acoustic prosodic features, the method is capable of estimating prominent words from continuous speech without labeled training data. The algorithm is evaluated by comparing model output to manually transcribed prominence labels on a Dutch and French speech corpus, showing performance levels close to supervised prominence classifiers operating on the same data. [ABSTRACT FROM AUTHOR]

Numerous studies have examined the acoustic correlates of sentential stress and its underlying linguistic functionality. However, the mechanism that connects stress cues to the listener's attentional processing has remained unclear. Also, the learnability versus innateness of stress perception has not been widely discussed. In this work, we introduce a novel perspective to the study of sentential stress and put forward the hypothesis that perceived sentence stress in speech is related to the unpredictability of prosodic features, thereby capturing the attention of the listener. As predictability is based on the statistical structure of the speech input, the hypothesis also suggests that stress perception is a result of general statistical learning mechanisms. To study this idea, computational simulations are performed where temporal prosodic trajectories are modeled with an n-gram model. Probabilities of the feature trajectories are subsequently evaluated on a set of novel utterances and compared to human perception of stress. The results show that the low-probability regions of F0 and energy trajectories are strongly correlated with stress perception, giving support to the idea that attention and unpredictability of sensory stimulus are mutually connected. [ABSTRACT FROM AUTHOR]

This study focuses on feature selection in paralinguistic analysis and presents recently developed supervised and unsupervised methods for feature subset selection and feature ranking. Using the standard k -nearest-neighbors (kNN) rule as the classification algorithm, the feature selection methods are evaluated individually and in different combinations in seven paralinguistic speaker trait classification tasks. In each analyzed data set, the overall number of features highly exceeds the number of data points available for training and evaluation, making a well-generalizing feature selection process extremely difficult. The performance of feature sets on the feature selection data is observed to be a poor indicator of their performance on unseen data. The studied feature selection methods clearly outperform a standard greedy hill-climbing selection algorithm by being more robust against overfitting. When the selection methods are suitably combined with each other, the performance in the classification task can be further improved. In general, it is shown that the use of automatic feature selection in paralinguistic analysis can be used to reduce the overall number of features to a fraction of the original feature set size while still achieving a comparable or even better performance than baseline support vector machine or random forest classifiers using the full feature set. The most typically selected features for recognition of speaker likability, intelligibility and five personality traits are also reported. [ABSTRACT FROM AUTHOR]

Highlights: [•] We model infants’ speech acquisition and emergence of the skill of imitation. [•] A virtual infant, LeVI, learns phonemes when its babbling is rewarded by a caregiver. [•] Imitation by the virtual caregiver teaches LeVI inverse auditory-to-phonetic mappings. [•] Imitations are not exact reproductions, and all 25 Finnish phonemes are learned. [•] LeVi learns to recognise the caregiver’s phonemes with an average accuracy of 88%. [Copyright &y& Elsevier]

Several psychoacoustic phenomena such as loudness perception, absolute thresholds of hearing, and perceptual grouping in time are affected by temporal integration of the signal in the auditory system. Similarly, the frequency resolution of the hearing system, often expressed in terms of critical bands, implies signal integration across neighboring frequencies. Although progress has been made in understanding the neurophysiological mechanisms behind these processes, the underlying reasons for the observed integration characteristics have remained poorly understood. The current work proposes that the temporal and spectral integration are a result of a system optimized for pattern detection from ecologically relevant acoustic inputs. This argument is supported by a simulation where the average time-frequency structure of speech that is derived from a large set of speech signals shows a good match to the time-frequency characteristics of the human auditory system. The results also suggest that the observed integration characteristics are learnable from acoustic inputs of the auditory environment using a Hebbian-like learning rule. [ABSTRACT FROM AUTHOR]

Abstract: This work reviews a number of existing computational studies concentrated on the question of how spoken language can be learned from continuous speech in the absence of linguistically or phonetically motivated background knowledge, a situation faced by human infants when they first attempt to learn their native language. Specifically, the focus is on how phonetic categories and word-like units can be acquired purely on the basis of the statistical structure of speech signals, possibly aided by some articulatory or visual constraints. The outcomes and shortcomings of the existing work are reflected onto findings from experimental and theoretical studies. Finally, some of the open questions and possible future research directions related to the computational models of language acquisition are discussed. [Copyright &y& Elsevier]

Abstract: An efficient method for weakly supervised pattern discovery and recognition from discrete categorical sequences is introduced. The method utilizes two parallel sources of data: categorical sequences carrying some temporal or spatial information and a set of labeled, but not exactly aligned, contextual events related to the sequences. From these inputs the method builds associative models able to describe systematically co-occurring structures in the input streams. The learned models, based on transitional probabilities of events observed at several different time lags, inherently segment and classify novel sequences into contextual categories. Learning and recognition processes are purely incremental and computationally cheap, making the approach suitable for on-line learning tasks. The capabilities of the algorithm are demonstrated in a keyword learning task from continuous infant-directed speech and a continuous speech recognition task operating at varying noise levels. [Copyright &y& Elsevier]

Abstract: Word segmentation from continuous speech is a difficult task that is faced by human infants when they start to learn their native language. Several studies indicate that infants might use several different cues to solve this problem, including intonation, linguistic stress, and transitional probabilities between subsequent speech sounds. In this work, a computational model for word segmentation and learning of primitive lexical items from continuous speech is presented. The model does not utilize any a priori linguistic or phonemic knowledge such as phones, phonemes or articulatory gestures, but computes transitional probabilities between atomic acoustic events in order to detect recurring patterns in speech. Experiments with the model show that word segmentation is possible without any knowledge of linguistically relevant structures, and that the learned ungrounded word models show a relatively high selectivity towards specific words or frequently co-occurring combinations of short words. [Copyright &y& Elsevier]

Abstract Purpose Method Result Conclusion The purpose of this study was to analyse the relationship between automatic vowel articulation index (aVAI) and direct magnitude estimation (DME) among speakers with Parkinson’s disease (PD) and healthy controls. We further analysed the potential of aVAI to serve as an objective measure of speech impairment in the clinical setting.Speech samples from native Finnish speakers were utilised. Expert raters utilised DME to scale the intelligibility of speech samples. aVAI scores for PD speakers and healthy control speakers were analysed in relationship to DME speech intelligibility ratings and, among PD speakers, disease stage utilising nonparametric statistical analysis.Mean DME intelligibility ratings were lower among PD speakers compared to healthy controls. Mean aVAI scores were nearly the same between speaker groups. DME intelligibility ratings and aVAI were strongly correlated within the PD speaker group. aVAI and DME intelligibility ratings were moderately correlated with disease stage as measured by the Hoehn and Yahr scale.aVAI was observed to be a promising tool for analysing vowel articulation in PD speakers. Further research is warranted on the application of aVAI as an objective measure of severity of speech impairment in the clinical setting, with varying patient populations and speech samples. [ABSTRACT FROM AUTHOR]

In order to study how early emotional experiences shape infant development, one approach is to analyze the emotional content of speech heard by infants, as captured by child-centered daylong recordings, and as analyzed by automatic speech emotion recognition (SER) systems. However, since large-scale daylong audio is initially unannotated and differs from typical speech corpora from controlled environments, there are no existing in-domain SER systems for the task. Based on existing literature, it is also unclear what is the best approach to deploy a SER system for a new domain. Consequently, in this study, we investigated alternative strategies for deploying a SER system for large-scale child-centered audio recordings from a neonatal hospital environment, comparing cross-corpus generalization, active learning (AL), and domain adaptation (DA) methods in the process. We first conducted simulations with existing emotion-labeled speech corpora to find the best strategy for SER system deployment. We then tested how the findings generalize to our new initially unannotated dataset. As a result, we found that the studied AL method provided overall the most consistent results, being less dependent on the specifics of the training corpora or speech features compared to the alternative methods. However, in situations without the possibility to annotate data, unsupervised DA proved to be the best approach. We also observed that deployment of a SER system for real-world daylong child-centered audio recordings achieved a SER performance level comparable to those reported in literature, and that the amount of human effort required for the system deployment was overall relatively modest. • Speech emotion recognition for real-life child-centered audio recordings. • Comparison of cross-corpus generalization, active learning, and domain adaptation. • The studied active learning method provided the most consistent results. • Using annotated data, active learning excelled. • Without annotated data, unsupervised domain adaptation excelled. [ABSTRACT FROM AUTHOR]

Purpose: The purpose of this study was to analyse the relationship between automatic vowel articulation index (aVAI) and direct magnitude estimation (DME) among speakers with Parkinson's disease (PD) and healthy controls. We further analysed the potential of aVAI to serve as an objective measure of speech impairment in the clinical setting., Method: Speech samples from native Finnish speakers were utilised. Expert raters utilised DME to scale the intelligibility of speech samples. aVAI scores for PD speakers and healthy control speakers were analysed in relationship to DME speech intelligibility ratings and, among PD speakers, disease stage utilising nonparametric statistical analysis., Result: Mean DME intelligibility ratings were lower among PD speakers compared to healthy controls. Mean aVAI scores were nearly the same between speaker groups. DME intelligibility ratings and aVAI were strongly correlated within the PD speaker group. aVAI and DME intelligibility ratings were moderately correlated with disease stage as measured by the Hoehn and Yahr scale., Conclusion: aVAI was observed to be a promising tool for analysing vowel articulation in PD speakers. Further research is warranted on the application of aVAI as an objective measure of severity of speech impairment in the clinical setting, with varying patient populations and speech samples.

Background: Early neurodevelopmental care and research are in urgent need of practical methods for quantitative assessment of early motor development. Here, performance of a wearable system in early motor assessment was validated and compared to developmental tracking of physical growth charts., Methods: Altogether 1358 h of spontaneous movement during 226 recording sessions in 116 infants (age 4-19 months) were analysed using a multisensor wearable system. A deep learning-based automatic pipeline quantified categories of infants' postures and movements at a time scale of seconds. Results from an archived cohort (dataset 1, N = 55 infants) recorded under partial supervision were compared to a validation cohort (dataset 2, N = 61) recorded at infants' homes by the parents. Aggregated recording-level measures including developmental age prediction (DAP) were used for comparison between cohorts. The motor growth was also compared with respective DAP estimates based on physical growth data (length, weight, and head circumference) obtained from a large cohort (N = 17,838 infants; age 4-18 months)., Findings: Age-specific distributions of posture and movement categories were highly similar between infant cohorts. The DAP scores correlated tightly with age, explaining 97-99% (94-99% CI 95) of the variance at the group average level, and 80-82% (72-88%) of the variance in the individual recordings. Both the average motor and the physical growth measures showed a very strong fit to their respective developmental models (R 2  = 0.99). However, single measurements showed more modality-dependent variation that was lowest for motor (σ = 1.4 [1.3-1.5 CI 95] months), length (σ = 1.5 months), and combined physical (σ = 1.5 months) measurements, and it was clearly higher for the weight (σ = 1.9 months) and head circumference (σ = 1.9 months) measurements. Longitudinal tracking showed clear individual trajectories, and its accuracy was comparable between motor and physical measures with longer measurement intervals., Interpretation: A quantified, transparent and explainable assessment of infants' motor performance is possible with a fully automated analysis pipeline, and the results replicate across independent cohorts from out-of-hospital recordings. A holistic assessment of motor development provides an accuracy that is comparable with the conventional physical growth measures. A quantitative measure of infants' motor development may directly support individual diagnostics and care, as well as facilitate clinical research as an outcome measure in early intervention trials., Funding: This work was supported by the Finnish Academy (314602, 335788, 335872, 332017, 343498), Finnish Pediatric Foundation (Lastentautiensäätiö), Aivosäätiö, Sigrid Jusélius Foundation, and HUS Children's Hospital/HUS diagnostic center research funds., Competing Interests: Declaration of interests E.Ilén is the owner and CEO of Planno. Planno provides the product and process R&D services in textile-based products. The other authors have no conflict of interest to report., (Copyright © 2023 The Author(s). Published by Elsevier B.V. All rights reserved.)

Infants' speech perception adapts to the phonemic categories of their native language, a process assumed to be driven by the distributional properties of speech. This study investigates whether deep neural networks (DNNs), the current state-of-the-art in distributional feature learning, are capable of learning phoneme-like representations of speech in an unsupervised manner. We trained DNNs with unlabeled and labeled speech and analyzed the activations of each layer with respect to the phones in the input segments. The analyses reveal that the emergence of phonemic invariance in DNNs is dependent on the availability of phonemic labeling of the input during the training. No increased phonemic selectivity of the hidden layers was observed in the purely unsupervised networks despite successful learning of low-dimensional representations for speech. This suggests that additional learning constraints or more sophisticated models are needed to account for the emergence of phone-like categories in distributional learning operating on natural speech.

A key feature of normal neonatal EEG at term age is interhemispheric synchrony (IHS), which refers to the temporal co-incidence of bursting across hemispheres during trace alternant EEG activity. The assessment of IHS in both clinical and scientific work relies on visual, qualitative EEG assessment without clearly quantifiable definitions. A quantitative measure, activation synchrony index (ASI), was recently shown to perform well as compared to visual assessments. The present study was set out to test whether IHS is stable enough for clinical use, and whether it could be an objective feature of EEG normality. We analyzed 31 neonatal EEG recordings that had been clinically classified as normal (n = 14) or abnormal (n = 17) using holistic, conventional visual criteria including amplitude, focal differences, qualitative synchrony, and focal abnormalities. We selected 20-min epochs of discontinuous background pattern. ASI values were computed separately for different channel pair combinations and window lengths to define them for the optimal ASI intraindividual stability. Finally, ROC curves were computed to find trade-offs related to compromised data lengths, a common challenge in neonatal EEG studies. Using the average of four consecutive 2.5-min epochs in the centro-occipital bipolar derivations gave ASI estimates that very accurately distinguished babies clinically classified as normal vs. abnormal. It was even possible to draw a cut-off limit (ASI~3.6) which correctly classified the EEGs in 97% of all cases. Finally, we showed that compromising the length of EEG segments from 20 to 5 min leads to increased variability in ASI-based classification. Our findings support the prior literature that IHS is an important feature of normal neonatal brain function. We show that ASI may provide diagnostic value even at individual level, which strongly supports its use in prospective clinical studies on neonatal EEG as well as in the feature set of upcoming EEG classifiers.