Babble Research Articles

Follow-up of expressive language and general development at 12, 18 and 36 months for children with no canonical babbling at 10 months

ABSTRACT Little is known about the developmental trajectories in children who do not use canonical babbling (CB) at 10 months. The aim was to examine how speech, language, cognitive and motor abilities developed in children without CB. For 15 children identified as not having CB, consonant production, expressive vocabulary and general development were assessed at 12, 18 and 36 months from audio–video recordings. Twelve (79%) children still lacked CB at 12 months. At 18 months, six (40%) had parent-reported expressive vocabulary results below the 10th percentile, and two (14%) did not use dental/alveolar stops. The percentage of consonants correct for the group was at the level of peers at 36 months (89%, SD 0.17), but the group had fewer established target consonants than age norms. Most children had a small expressive vocabulary in comparison with Swedish age norms for younger children as well as with age-matched norms for other Nordic languages. The general development (Bayley-III) for three children (21%) who did not use speech was 1–2 SD below the average range in at least one domain (cognitive, language or motor), but the results for the whole group were within the average range. In conclusion, the heterogeneity in early consonant development in the study group resembles that of children in clinical groups with known risk for speech and language difficulties and the expressive vocabulary resembles that of children with delayed expressive vocabulary. For about one-fifth of the children, the absence of canonical babbling at age 10 months could be seen as an early sign of a more comprehensive developmental disability.

Baby Cry Sound Detection: A Comparison of Mel Spectrogram Image on Convolutional Neural Network Models

Baby cries contain patterns that indicate their needs, such as pain, hunger, discomfort, colic, or fatigue. This study explores the use of Convolutional Neural Network (CNN) architectures for classifying baby cries using Mel Spectrogram images. The primary objective of this research is to compare the effectiveness of various CNN architectures such as VGG-16, VGG-19, LeNet-5, AlexNet, ResNet-50, and ResNet-152 in detecting baby needs based on their cries. The datasets used include the Donate-a-Cry Corpus and Dunstan Baby Language. The results show that AlexNet achieved the best performance with an accuracy of 84.78% on the Donate-a-Cry Corpus dataset and 72.73% on the Dunstan Baby Language dataset. Other models like ResNet-50 and LeNet-5 also demonstrated good performance although their computational efficiency varied, while VGG-16 and VGG-19 exhibited lower performance. This research provides significant contributions to the understanding and application of CNN models for baby cry classification. Practical implications include the development of baby cry detection applications that can assist parents and healthcare provide.

Contingency enables the formation of social expectations about an artificial agent.

What environmental regularities support infant communicative learning from social interactions? We propose that infants allocate their attention toward and learn from external events that are contingent on their own behaviors. We tested the robustness of the influence of contingency on communicative learning by using a non-biological stimulus, a remote-controlled car, as the social partner. The car approached infants and produced a speech sound either contingently to infants' vocalizations or on a yoked schedule. Two additional groups had an unfamiliar human experimenter as their social partner in contingent and yoked control conditions. We assessed whether infants formed expectations about their partner's responsiveness to their vocalizations. Expectations made based on contingent responsiveness would support the role of contingency in promoting plasticity in early communicative learning. Infants across all conditions increased their vocalization rates when their partner paused in responding, suggesting that they expected their vocalizations to influence their partners' behavior. Infants vocalized significantly more to the social partner than their caregiver if they received contingent rather than yoked responses from the social partner, regardless of if the partner was a human or non-biological agent. Contingent responses to prelinguistic vocalizations facilitated the formation of expectations for interactivity of social partners.

Vocal and early speech development in Korean-acquiring children with hearing loss and typical hearing

ABSTRACT This study investigated the vocal and early speech development of Korean-acquiring children with hearing loss (HL) who underwent early auditory amplification compared to their typical hearing (TH) counterparts. The research focused on phonological characteristics of child vocalisation based on samples collected from naturalistic home environments. One-day home recordings using a Language ENvironment Analysis (LENA) recorder were obtained from 6 children with HL and 12 children with TH who ranged from 17 to 23 months of age in Korean monolingual environments. Child volubility, canonical babbling ratio (CBR), consonant distributions, and utterance structures of vocalisations were evaluated through qualitative and quantitative analyses of vocalisation samples collected from LENA recordings. The findings revealed that children with HL displayed comparable vocalisation levels to children with TH, with no significant differences in volubility and CBR. In consonant and utterance shape inventories, noticeable quantitative and qualitative differences were observed between children with HL and those with TH. The study also suggested both universal and language-specific production patterns, revealing the early effects of ambient language on consonant distributions and utterance structures within their vocalisation repertoire. This study emphasised the role of auditory input and the importance of early auditory amplification to support speech development in children with HL.

Canonical babbling trajectories across the first year of life in autism and typical development.

Our study examined how babies develop their ability to talk to help identify early signs of autism. We looked at babies' production of babbling with mature syllables across the first year of life. Babies usually start producing mature babbling at 7 months of age before they say their first words. Some studies have suggested that babies who are later diagnosed with autism produce this kind of babbling less frequently in their first year of life, but other studies have shown complicated outcomes. In this new study, we followed 44 autistic babies and compared them to 127 typically developing babies. We recorded the babies once every month, all day long, from the time that they were born until they were around 13 months old. Then, we studied their mature babbling from segments of these recordings. We found that the rate at which babies used mature babbling was lower in boys with autism, and higher in girls with autism, compared to babies without autism. This research helps us understand how babies with autism learn to talk. It also raises important questions about differences between boys and girls with autism. Our study can help us improve how scientists and clinicians can identify autism earlier, which could lead to better communication supports for autistic children and their families.

The Predictability of Naturalistic Evaluation of All-Day Recordings for Speech and Language Development.

The study aimed to investigate the predictive potential of language environment and vocal development status measures obtained through integrated analysis of Language ENvironment Analysis (LENA) recordings during the prelinguistic stage for subsequent speech and language development in Korean-acquiring children. Specifically, this study explored whether measures from both LENA-automated analysis and human coding at 6-8 months and 12-14 months of age predict vocabulary and phonological development at 18-20 months. One-day home recordings from 20 children were collected using a LENA recorder at 6-8 months, 12-14 months, and 18-20 months. Both LENA-automated measures and measures from human coding were obtained from recordings at 6-8 months and 12-14 months. The number of different words, consonant inventory, and utterance structure inventory were identified from recordings of 18-20 months. Correlation and multiple regression analyses were performed to investigate whether measures related to early language environment and child vocalization at 6-8 months and 12-14 months were predictive of vocabulary and phonological measures at 18-20 months. The results showed that the two main LENA-automated measures, conversational turn count (CTC) and child vocalization count, were positively correlated with all vocabulary and phonological measures at 18-20 months. Multiple regression analysis revealed that CTC during the prelinguistic stages was the most significant predictor of a number of different words, consonant inventory, and utterance structure inventory at 18-20 months. Also, adult word count in LENA-automated measures, child-directed speech ratio, and canonical babbling ratio measured by human coding significantly predicted some vocabulary and phonological measures at 18-20 months. This study highlights the multifaceted nature of language acquisition and collectively emphasizes the value of considering both quantitative and qualitative aspects of language input to understand early language development in children.

Developmental cascades of vocal turn-taking connect prelinguistic vocalizing with early language

Conversational turn-taking is ubiquitously found in caregiver-infant interactions, and robustly predictive of infant communicative development. Over the first year, infants take quick adult-like vocal turns with caregivers. Many studies have documented the consistency of caregiver responsiveness and its influence on infant rapid language growth. However, few have examined how caregiver responsiveness facilitates extended vocal turn-taking in real-time with infants over the first year. The influence of prelinguistic vocal turn-taking on the emergence of language has also been under-investigated. We analyzed free-play sessions of 30 caregivers and their infants at both 5 and 10 months, and obtained infant language outcomes at 18 months. We examined the developmental consistency (group-level continuity and dyad-order stability) and change of infant volubility, caregiver responses to babbling in vocal, non-vocal and multimodal modalities, and the influence of modality on caregiver-infant vocal turn-taking. Caregiver contingent responsiveness to infant babbling at 5 months predicted vocal turn-taking at 10 months. Developmental increases in prelinguistic vocalizing and vocal turn-taking from 5 to 10 months predicted infant language outcomes at 18 months. At both 5 and 10 months, caregiver vocal responses were more effective in extending turn-taking than non-vocal or multimodal responses. In summary, prelinguistic vocal turn-taking, facilitated by caregiver vocal responsiveness, is positively related to the emergence of early language.

Infant cry classification using an efficient graph structure and attention-based model

Crying serves as the primary means through which infants communicate, presenting a significant challenge for new parents in understanding its underlying causes. This study aims to classify infant cries to ascertain the reasons behind their distress. In this paper, an efficient graph structure based on multi-dimensional hybrid features is proposed. Firstly, infant cries are processed to extract various speech features, such as spectrogram, mel-scaled spectrogram, MFCC, and others. These speech features are then combined across multiple dimensions to better utilize the information in the cries. Additionally, in order to better classify the efficient graph structure, a local-to-global convolutional neural network (AlgNet) based on convolutional neural networks and attention mechanisms is proposed. The experimental results demonstrate that the use of the efficient graph structure improved the accuracy by an average of 8.01% compared to using standalone speech features, and the AlgNet model achieved an average accuracy improvement of 5.62% compared to traditional deep learning models. Experiments were conducted using the Dunstan baby language, Donate a cry, and baby cry datasets with accuracy rates of 87.78%, 93.83%, and 93.14% respectively.

A Machine Learning Approach to Classify Biomedical Acoustic Features for Baby Cries

Communication is imperative for living beings for exchanging information. But for newborns, the only way of communicating with the world is through crying, and it is the only medium through which caregivers can know about the needs of their children. Timely addressing baby cries is very important so that the child is relieved at the earliest. It has been a challenge, especially for new parents. The literature says newborn babies use The Dustan Baby Language to communicate. According to this language, there are five words to understand a baby’s needs, which are “Neh” (hungry), “Eh” (burp is needed), “Owh/Oah” (fatigue), “Eair/Eargghh” (cramps), “Heh” (feel hot or wet, physical discomfort). This research aims to develop a model for recognizing baby cries and distinguishing between different kinds of baby cries. Here we more broadly focus on whether the infant is in pain due to hunger or discomfort. The study proposes a comparative approach using four classification models: random forest, support vector machine, logistic regression, and decision tree. These algorithms learn from the spectral features: chroma_stft, spectral_centroid, bandwidth, spectral_rolloff, mel-frequency cepstral coefficients, linear predictive coding, res, zero_crossing_rate extracted from the infant cry. The support vector machine model outperforms other classifiers for correctly classifying infant cries.

Understanding speech and language in tuberous sclerosis complex.

Tuberous Sclerosis Complex (TSC), is a neurocutaneous disorder, associated with a high prevalence of autism spectrum disorder (ASD; ∼50% of individuals). As TSC is a leading cause of syndromic ASD, understanding language development in this population would not only be important for individuals with TSC but may also have implications for those with other causes of syndromic and idiopathic ASD. In this mini review, we consider what is known about language development in this population and how speech and language in TSC are related to ASD. Although up to 70% of individuals with TSC report language difficulties, much of the limited research to date on language in TSC has been based on summary scores from standardized assessments. Missing is a detailed understanding of the mechanisms driving speech and language in TSC and how they relate to ASD. Here, we review recent work suggesting that canonical babbling and volubility-two precursors of language development that predict the emergence of speech and are delayed in infants with idiopathic ASD-are also delayed in infants with TSC. We then look to the broader literature on language development to identify other early precursors of language development that tend to be delayed in children with autism as a guide for future research on speech and language in TSC. We argue that vocal turn-taking, shared attention, and fast mapping are three such skills that can provide important information about how speech and language develop in TSC and where potential delays come from. The overall goal of this line of research is to not only illuminate the trajectory of language in TSC with and without ASD, but to ultimately find strategies for earlier recognition and treatment of the pervasive language difficulties in this population.

Parental Responses to Prelinguistic Vocalizations of Very Young Children with Typical Development: A Systematic Review and Meta-Analysis

Parental contingent responses (PCRs) to infant’s vocalization are related to early communication and language development. This study identified the parental response patterns to infant vocalizations in the interaction between infants and their parents, the pattern of infant vocalizations, and parental language-expectant responses in the PCRs through a systemic literature review and meta-analysis. According to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), seven studies were met Participants, Intervention, Comparison, Outcomes, and Study design (PICOS) criteria from 8 databases. The quality of studies evaluated risk of bias and effect sizes were calculated using Hedges' <i>g</i>. As a result of systematic review, the selected studies had an overall low risk of bias, and the subjects of all studies were infants who are within 14 months and mothers. As a result of meta-analysis, PCRs to infant vocalizations were significantly higher than parental non-contingent responses. Language-expectant responses appeared significantly more than language non-expectant responses in PCRs. This study showed that parents are more likely to respond verbally to their infants’ vocalizations with their infants, suggesting the significant role is very important in early interaction. These finding may provide foundation material to guide early intervention and parent coaching for infants at risk for communication disorders.

An ensemble multi-stream classifier for infant needs detection

In this paper, we propose a novel multi-stream video classifier for infant needs detection. The proposed system is an ensemble-based system that combines several machine learning to improve the overall result of the state-of-the-art algorithms. It is a multi-stream in the sense that it combines the output predictions of both audio and images of infants from every single classifier employed in the system for a unified result. This produces better performance and results compared to the previous other research techniques, which relied on only one of these modalities. For training and testing the proposed system, from the Dunstan Baby Language video collection, we built three separate datasets for videos, images, and sounds encompassing the five primary infant needs that require predicting. These are: hunger, have wind, uncomfortable (require diaper change), wants to burp or tired, with a total of 3348 samples. We used four different ensemble algorithms for the best reachable performance. The proposed algorithm improves the overall accuracies of each single classifier from a low of 51% to a high of 99%. The proposed method also improves the accuracy of the classification process by about 9% compared to the state-of-the-art approaches, which was 90%.

Life of p: A consonant older than speech.

Which sounds composed the first spoken languages? Archetypal sounds are not phylogenetically or archeologically recoverable, but comparative linguistics and primatology provide an alternative approach. Labial articulations are the most common speech sound, being virtually universal across the world's languages. Of all labials, the plosive 'p' sound, as in 'Pablo Picasso', transcribed /p/, is the most predominant voiceless sound globally and one of the first sounds to emerge in human infant canonical babbling. Global omnipresence and ontogenetic precocity imply that /p/-like sounds could predate the first major linguistic diversification event(s) in humans. Indeed, great ape vocal data support this view, namely, the only cultural sound shared across all great ape genera is articulatorily homologous to a rolling or trilled /p/, the 'raspberry'. /p/-like labial sounds represent an 'articulatory attractor' among living hominids and are likely among the oldest phonological features to have ever emerged in linguistic systems.

Marginal and Canonical Babbling in 10 Infants at Risk for Cerebral Palsy.

This study is a preliminary quantification and characterization of the development of marginal and canonical syllable patterns in 10 infants at risk for cerebral palsy (CP). We calculated marginal and canonical babbling ratios from parent-infant laboratory recordings of 10 infants at two time points, approximately 13 and 16 months of age. The frequency and diversity of labial, coronal, and velar types of marginal and canonical syllables were also examined. Differences across three outcome groups were compared: infants later diagnosed with CP (n = 3, CP group), risk of CP due to ongoing gross motor delays (n = 4, risk group), and current typically developing status with resolved gross motor delays (n = 3, TDx group). Performance on the Mullen Scales was included for perspective on cognitive development. Higher marginal syllable ratios were observed in the CP and risk groups than the TDx group. An increasing canonical syllable ratio across the two ages was consistently observed in the TDx group. The TDx group produced a greater frequency and diversity of canonical syllable types than the risk and CP groups, and of marginal syllable types than the CP group. This study offers preliminary support for the possibility that speech motor impairment in infants with CP have the potential to be observed and quantified early in vocal development prior to the expected onset of first words. Prolonged rates of marginal syllable forms may be suggestive of speech motor impairment; however, additional longitudinal outcome data over a longer time course and a larger sample of infants are needed to provide further support for this possibility.

Phonetic transcription of spontaneous children's speech with the aid of software: a systematic review

The aim of the study was to identify, synthesize and classify the software currently available that can help in the task of phonetic transcription of the spontaneous speech of pre-school children to evaluate the development of children's language. A systematic review was performed for articles published, for the 10-year period (June 2010 to June 2020), without restrictions as to location and language, using the Cochrane, Pubmed and Web of Science databases. The terms used in the search strategies were "phonological", "phonetic", "transcription", "computer" and "software". The studies were selected by two independent reviewers using pre-defined search strategies. In the initial search, after the exclusion of duplicates, 534 articles were found. By reading their titles and abstracts, 46 articles related to the theme were left, which were then read in full. After reading, 24 articles were included in the study. The results revealed a total of seven software available for the phonetic transcription of spontaneous speech from preschoolers used for different analyses: LENA and Timestamper (for babbling and pre-linguistic vocalizations), ELAN (for gestural communication, extralinguistic elements and the situational context), Phon (for phonetic and phonological analyses), CLAN and SALT (for morphosyntactic, grammatical and semantic aspects) and Praat (for acoustic measurements). Through this systematic review, it can be concluded that there are advantages to using software for phonetic transcription, sample storage, and child language analysis, especially concerning standardization and reliability for spontaneous speech samples. Phonetic transcription still relies on the ability and subjectivity of a human transcriber. The tools found in the software provide support to facilitate using phonetic symbols, audio segmentation and pairing to writing, and analysis of speech data.

Babbling in extremely premature infants at 12 months corrected age

ABSTRACT Babbling is an important precursor to speech in infancy, and deviations from the typical babbling development can predict later difficulties in speech, language, and communication. This cross-sectional study aimed to investigate babbling and early speech in Swedish extremely premature infants. Samples of babbling were collected from 20 extremely premature infants (EPT group) at the corrected age of 12 months. Data collection was home-based and consisted of an audio-video recording of each infant playing with a parent. Presence of canonical babbling (CB), and three oral stop variables distinctive of typical babbling, and consonant inventory were assessed. The assessment was performed during a standardised observation of babbling. Data from the EPT group were compared to previously collected data of a reference group of 20 10-month-old infants without known medical diagnoses. The results showed that the EPT group had a lower proportion of infants producing CB, and that they used a significantly smaller consonant inventory compared to the reference group. Although not statistically significant, oral stops were less frequently found in the EPT group. The findings of a restricted consonant inventory and low proportion of CB in the EPT group are not surprising considering that the group has been found to be at risk of speech and language delay in toddlerhood. Still, further research is needed to explore whether babbling at 12 months can predict speech and language skills at an older age in extremely premature infants.

Phonetic transcription of spontaneous children's speech with the aid of software: a systematic review

The aim of the study was to identify, synthesize and classify the software currently available that can help in the task of phonetic transcription of the spontaneous speech of pre-school children to evaluate the development of children's language. A systematic review was performed for articles published, for the 10-year period (June 2010 to June 2020), without restrictions as to location and language, using the Cochrane, Pubmed and Web of Science databases. The terms used in the search strategies were "phonological", "phonetic", "transcription", "computer" and "software". The studies were selected by two independent reviewers using pre-defined search strategies. In the initial search, after the exclusion of duplicates, 534 articles were found. By reading their titles and abstracts, 46 articles related to the theme were left, which were then read in full. After reading, 24 articles were included in the study. The results revealed a total of seven software available for the phonetic transcription of spontaneous speech from preschoolers used for different analyses: LENA and Timestamper (for babbling and pre-linguistic vocalizations), ELAN (for gestural communication, extralinguistic elements and the situational context), Phon (for phonetic and phonological analyses), CLAN and SALT (for morphosyntactic, grammatical and semantic aspects) and Praat (for acoustic measurements). Through this systematic review, it can be concluded that there are advantages to using software for phonetic transcription, sample storage, and child language analysis, especially concerning standardization and reliability for spontaneous speech samples. Phonetic transcription still relies on the ability and subjectivity of a human transcriber. The tools found in the software provide support to facilitate using phonetic symbols, audio segmentation and pairing to writing, and analysis of speech data.

Perspectives on the origin of language: Infants vocalize most during independent vocal play but produce their most speech-like vocalizations during turn taking.

A growing body of research emphasizes both endogenous and social motivations in human vocal development. Our own efforts seek to establish an evolutionary and developmental perspective on the existence and usage of speech-like vocalizations ("protophones") in the first year of life. We evaluated the relative occurrence of protophones in 40 typically developing infants across the second-half year based on longitudinal all-day recordings. Infants showed strong endogenous motivation to vocalize, producing vastly more protophones during independent vocal exploration and play than during vocal turn taking. Both periods of vocal play and periods of turn-taking corresponded to elevated levels of the most advanced protophones (canonical babbling) relative to periods without vocal play or without turn-taking. Notably, periods of turn taking showed even more canonical babbling than periods of vocal play. We conclude that endogenous motivation drives infants' tendencies to explore and display a great number of speech-like vocalizations, but that social interaction drives the production of the most speech-like forms. The results inform our previously published proposal that the human infant has been naturally selected to explore protophone production and that the exploratory inclination in our hominin ancestors formed a foundation for language.

Fundamental Frequency Contour (Melody) of Infant Vocalizations across the First Year

Introduction: The fundamental frequency contour (melody) of cry and non-cry utterances becomes more complex with age. However, there is a lack of longitudinal analyses of melody development during the first year of life. Objective: The aim of the study was to longitudinally analyze melody development in typical vocalization types across the first 12 months of life. The aim was twofold: (1) to answer the question whether melody becomes more complex in all vocalization types with age and (2) to characterize complex patterns in more detail. Methods: Repeatedly recorded vocalizations (n = 10,988) of 10 healthy infants (6 female) over their first year of life were analyzed using frequency spectrograms and fundamental frequency (f0) analyses (PRAAT). Melody complexity analysis was performed using specific in-lab software (CDAP, pw-project) in a final subset of 9,237 utterances that contained noise-free, undisturbed contours. Generalized mixed linear models were used to analyze age and vocalization type effects on melody complexity. Results: The vocalization repertoire showed a higher proportion of complex melodies from the second month onward. The age effect was significant, but no difference was found in melody complexity between cry and non-cry vocalizations across the first 6 months. From month 7–12, there was a further significant increase in complex structures only in canonical babbling not in marginal babbling. Melody segmentations by laryngeal constrictions prevailed among complex shapes. Conclusion: The study demonstrated the regularity of melody development in different vocalization types throughout the first year of life. In terms of prosodic features of infant sounds, melody contour is of primary importance, and further studies are required that also include infants at risk for language development.

Infant vocalizations elicit simplified speech in childcare.

Infant babbling has an important social function in promoting early language development by attracting caregiver attention and prompting parents' contingent, simplified speech, which is more learnable for infants. Here, we demonstrate that prelinguistic infant vocalizations also create learning opportunities for infants in childcare settings by eliciting simplified and more learnable linguistic information during teacher-infant interactions. We compared the rates and complexity of contingent and non-contingent verbal interactions of 34 childcare teachers during a one-on-one free play interaction with a familiar infant under their care (M=12.6months old). As compared to non-contingent utterances, teachers' contingent utterances included fewer unique words, a higher proportion of single-word responses, and a shorter mean length of utterances. Teachers did not change their response length based on infants' syllable type and were equally likely to respond to vowels and consonant-vowel vocalizations. Sources of individual differences in the simplification effect related to infant behaviors and teacher characteristics are discussed. The results parallel previous findings demonstrating the simplification effect in parent-infant interactions. That teachers also show this simplification effect when responding to infant vocalizations suggests the power of infant prelinguistic vocalizations for organizing caregiver attention in various settings to elicit simplified, learnable language.