Types Of Vocalizations Research Articles

Foundations of Vocal Category Development in Autistic Infants.

The present study comparedthe infant's tendency in the first yearof life to produce clusters of particular vocal types (squeals, vocants, and growls) in typically developing (TD) and autistic infants. Vocal clustering provides evidence of vocal category formation and may establish a foundation for speech development. Specifically, we compared the extent of vocal clustering across outcome groups and age groups. We also examined the associations between the extent of vocal clustering and later outcomes at 2years within the autistic group. Randomly selected 5-min segments (27,153 5-min segments total) from 1293 all-day home recordings from 103 TD infants and 44 autistic infants across the first year were humancoded (about 9.75h of data coded per infant on average) to derive vocal clustering patterns. Fisher's exact tests were used to compare the occurrence of squeals versus vocants, as well as growls versus vocants, across coded segments. Infants in both groups demonstrated clear clustering patterns of squeals and growls across all age groups. The extent of vocal clustering in the autistic group did not correlate significantly with later language, repetitive behavior, or autism severity outcomes. These findings highlight the robustness of the systematic production of vocal categories across the first year of life. The similarity of the clustering patterns in the TD and autistic groups suggests that vocal category formation through active infant vocal exploration is a robust feature of early speech development.

BovineTalk: machine learning for vocalization analysis of dairy cattle under the negative affective state of isolation.

There is a critical need to develop and validate non-invasive animal-based indicators of affective states in livestock species, in order to integrate them into on-farm assessment protocols, potentially via the use of precision livestock farming (PLF) tools. One such promising approach is the use of vocal indicators. The acoustic structure of vocalizations and their functions were extensively studied in important livestock species, such as pigs, horses, poultry, and goats, yet cattle remain understudied in this context to date. Cows were shown to produce two types of vocalizations: low-frequency calls (LF), produced with the mouth closed, or partially closed, for close distance contacts, and open mouth emitted high-frequency calls (HF), produced for long-distance communication, with the latter considered to be largely associated with negative affective states. Moreover, cattle vocalizations were shown to contain information on individuality across a wide range of contexts, both negative and positive. Nowadays, dairy cows are facing a series of negative challenges and stressors in a typical production cycle, making vocalizations during negative affective states of special interest for research. One contribution of this study is providing the largest to date pre-processed (clean from noises) dataset of lactating adult multiparous dairy cows during negative affective states induced by visual isolation challenges. Here, we present two computational frameworks-deep learning based and explainable machine learning based, to classify high and low-frequency cattle calls and individual cow voice recognition. Our models in these two frameworks reached 87.2 and 89.4% accuracy for LF and HF classification, with 68.9 and 72.5% accuracy rates for the cow individual identification, respectively.

Midbrain neurons important for the production of mouse ultrasonic vocalizations are not required for distress calls

A fundamental feature of vocal communication is that animals produce vocalizations with different acoustic features in different behavioral contexts (contact calls, territorial calls, courtship calls, etc.). The midbrain periaqueductal gray (PAG) is a key region that regulates vocal production, and artificial activation of the PAG can elicit the production of multiple species-typical vocalization types.1,2,3,4,5,6,7,8,9 How PAG circuits are organized to regulate the production of different vocalization types remains unknown. On the one hand, studies have found that partial PAG lesions abolish the production of some vocalization types while leaving others intact,3,8,10,11 suggesting that different populations of PAG neurons might control the production of different vocalization types. On the other hand, electrophysiological recordings have revealed individual PAG neurons that increase their activity during the production of multiple vocalization types,12,13,14 suggesting that some PAG neurons may regulate the production of more than one vocalization type. To test whether a single population of midbrain neurons regulates the production of different vocalization types, we applied intersectional methods to selectively ablate a population of midbrain neurons important for the production of ultrasonic vocalizations (USVs) in mice. We find that, although ablation of these PAG-USV neurons blocks USV production in both males and females, these neurons are not required for the production of distress calls. Our findings suggest that distinct populations of midbrain neurons control the production of different vocalization types.

Aquatic Soundscape Recordings Reveal Diverse Vocalizations and Nocturnal Activity of an Endangered Frog.

AbstractAutonomous sensors provide opportunities to observe organisms across spatial and temporal scales that humans cannot directly observe. By processing large data streams from autonomous sensors with deep learning methods, researchers can make novel and important natural history discoveries. In this study, we combine automated acoustic monitoring with deep learning models to observe breeding-associated activity in the endangered Sierra Nevada yellow-legged frog (Rana sierrae), a behavior that current surveys do not measure. By deploying inexpensive hydrophones and developing a deep learning model to recognize breeding-associated vocalizations, we discover three undocumented R. sierrae vocalization types and find an unexpected temporal pattern of nocturnal breeding-associated vocal activity. This study exemplifies how the combination of autonomous sensor data and deep learning can shed new light on species' natural history, especially during times or in locations where human observation is limited or impossible.

제2언어 습득에서의 전이 현상 연구

We are living in an era of acquiring more than two languages in an internationalized and multicultural society. As a second language in the international community, standard Chinese is still one of the languages that receives a lot of attention. This study examined the influence of the second language on the first language in the process of learning Chinese as a second language. Specifically, an experiment was conducted on the calculation of closed sounds in standard Chinese and Korean for native Chinese speakers and Korean students studying in China to compare and analyze the acoustic characteristics (H1-H2) of the vocal type. As a result of the analysis, significant changes were found in the realization of H1-H2 of the Korean stop, which is the native language, by a group of high-level speakers using a second language as a standard Chinese. These results of this study are significant as another basis for supporting the impact of the language environment faced by bilingual users on the output of second languages and the backward transfer phenomenon of their first language during second language acquisition.

Analysis Of Factors Affecting Bird Identification Mechanisms and Suggestions for Environmental Protection

Recognition, happening between species, kin, neighbors, and mates, is the basis of most animal interactions such as mating, food provisioning, and territory protection. The identification accuracy of conspecifics is facilitated by some factors but impeded by others, such as the cost of acquiring and memorizing social information. Avians primarily use vocal and visual identity signals to recognize other individuals, such as the general song characteristics or spot patterns on the feathers. By analyzing the identification context, signature production, and signal perception and response, the article reviews recent works to recommend environmental conservation approaches from the perspective of bird identification. Factors including sex, behavioral context, vocalization types, distance, and cognitive cost are demonstrated in their influence on recognition, and the lack of research on visual recognition does not indicate the more adaptive and complex nature of vocal recognition. Furthermore, the review encourages reducing anthropogenic effects on noise, light, and climate; avoiding behavioral disturbance; and restoring natural habitats. The review provides insights into the current state of individual animal recognition research by organizing the main findings of recent studies.

The vowel /a/ as the main portal to humanity’s language and culture faculties

The vowel /a/ is regarded here as the initial sound, based on earlier vowel-like vocalization in humans, especially the neonate cry. This particular type of vocalization marks the true beginning of human language in the ontological perspective. Its presence is absolutely fundamental for the generation and maintenance of oxygen-based language and culture complex. All of human life is conducted in the human auditive world of organization based on the air (the aerial condition).

Analysis of the Art and Emotional Skills of College Vocal Singing in the Age of Big Data

Abstract In this paper, the extracted acoustic features are processed using the MFCC method to classify the sound recognition. The linear spectrum of the sound is mapped onto the Mel FPP coefficients and Mel nonlinear spectrum, respectively, and the MFCC is obtained by cepstrum processing. The Mel FPP coefficients and Gaussian mixture model are combined to create the acoustic model. On this basis, the art and emotional skills of vocal singing are explored, and simulation and empirical experiments are set up to analyze the effect of the model constructed in this paper. The experimental results show that the model constructed in this paper tends to stabilize when the number of iterations is 2, the final accuracy rate stabilizes near 0.9, and the model is effective. Different vocal types are recognized by the model constructed in this paper, and the recognition rate of emotional expression of all vocal types is above 0.7, and the model recognition rate is high. The accuracy and recall classification for emotional expression techniques are 0.73 and 0.79, resulting in a more balanced evaluation overall.

Context effects on duration, fundamental frequency, and intonation in human-directed domestic cat meows

In this study, we investigated the prosody of domestic cat meows produced in different contexts. Prosodic cues – i.e., variation in intonation, duration, voice quality and fundamental frequency – in humans as well as in nonhuman animals carry information about idiosyncratic traits of the signaller, including sex, age, and physical and mental state. The duration, fundamental frequency (F0) and intonation in a sample of 969 meows recorded in seven different contexts (i.e., cuddle, door, food, greeting, lifting, play, cat carrier) were analysed using linear mixed effects regression and generalized additive models. In this, we controlled for cat age and sex, as meows produced by old cats had lower mean F0 than those produced by young cats, and female cats produced meows with higher mean F0 than male cats. We found significant effects of context on duration and mean F0, but not on F0 range. Furthermore, the results showed that the intonation of meows produced by cats in a cat carrier displayed a falling pattern, while that of meows produced in cuddle and door contexts was relatively level, and that of meows produced in the other contexts consisted of combinations of rising and falling. The average slope of meows produced in cat carrier and play contexts was negative, while that of meows produced in the other contexts was positive. We argue that this prosodic variation reflects the cats’ mental or emotional state, because of valence and arousal differences associated with the various contexts that were included in the study. Further studies will need to confirm this. In addition, we also plan additional analyses of spectral and voice quality parameters in meows and other cat vocalisation types.

Singing silver‐haired bats (Lasionycteris noctivagans)

AbstractCharacterizing sounds produced by animals can lead to better understanding of their behavioral ecology and conservation. While considerable focus has been on signals used by bats for echolocation, there has been less emphasis on nonecholocation sounds. We describe songs (i.e., acoustic vocalizations with distinctive syllable types in series or in complex motifs) produced by silver‐haired bat (Lasionycteris noctivagans). Songs, characterized by a sequence (song phrase) of 3 distinct vocalization types, were confirmed by observing free‐flying, silver‐haired bats at mine hibernacula in British Columbia, Canada. The song patterns were relatively consistent with each song phrase consisting of a lead call, followed by a droplet call, and finishing with a series of multiple chirp calls. The function of the songs is unknown, however, as other bat species produce songs for mating, we propose silver‐haired bat songs may similarly be associated with courtship or mating. Alternative functions cannot be ruled out, particularly because we recorded some songs outside of the accepted mating period. Other research has determined peak mating of silver‐haired bats occurs in fall, and spring mating has been documented. Here we additionally provide evidence of winter mating in British Columbia. The proportion of silver‐haired bat songs recorded relative to echolocation recordings varied across locations and seasons. While we recorded songs in all months of the year, more than half of the songs were produced during winter, and 93.4% (of 1,857) were produced outside of summer months. Song production in summer could be associated with other behaviors such as learning or practice, establishing or maintaining social bonds, or male‐male competition. To provide landscape and temporal context, we summarize acoustic datasets from numerous locations in western North America where recordings were made between 2005 and 2022.

139 Genomic-Based Genetic Parameters of Behavioral Ethogram Traits in Lactating Sows Under Heat Stress Conditions and Their Relationship with Direct Indicators of Heat Tolerance

Abstract Increased global temperatures pose challenges to livestock agriculture. Domestic swine are particularly sensitive to hotter environments due to their inability to sweat, litter bearing nature, difficulty in expressing natural cooling behaviors, and intense genetic selection for production traits. Since animal welfare is central to livestock production as a matter vital to efficiency as much as ethics, developing novel traits that may be genetically related to heat tolerance may assist in improving and sustaining animal welfare. The objective of this study was to develop novel traits of behavior using an ethogram designed for efficiency in production settings, estimate the variance components, and correlate these traits with indicators of heat tolerance and maternal ability. This study considered 1,312 individual Landrace-Large White cross production sows between 6 and 19 days of lactation. Animals were housed in farrowing crates in a commercial setting in southern North Carolina in June and July of 2021, an environment susceptible to potential heat stress. These sows were subjected to a standardized procedure involving a person removing hair from both hips using an electric clipper while an observer was positioned in front of the sow to record behaviors. Scores were later applied to the variations in behavior both for changes in positions and types of vocalizations displayed by the animals during handling. Genotypes were obtained from hair follicles and captured by a 50K porcine SNP panel. Variance components for novel behavior traits were obtained using the BLUPF90+ programs and genomic relationship matrix to estimate genetic correlations of behavior with several indicators of heat tolerance and maternal ability. Heritabilities for behavior traits ranged from 0.131 for vocalization score to 0.072 for shaving time, while heritabilities for heat tolerance and maternal ability ranged from 0.15 to 0.03 for vaginal temperature and respiration efficiency, respectively. The maternal ability traits were based on the single current litter and had the lowest heritabilities ranging from 0.101 to 0.04 for number born alive and number of mummies, respectively. High and positive genetic correlations indicate a potential for coupled selection for both fewer mummified piglets and a greater number of liveborn using the behavior trait of vocalization score. These correlations between favorable maternal traits and behavior traits suggest the potential that one or more complex biological pathways exist that share relevancy between these traits. However, genetic correlations between the behavior traits and heat tolerance indicators were low to negligible, most with a magnitude falling below 0.10, revealing the need for other traits to be used as selection parameters for heat tolerance. Further studies may be able to capture additional environmental effects to improve models of behavior and create indirect measures more obtainable than direct measures of heat tolerance.

Simultaneous recording of ultrasonic vocalizations and sniffing from socially interacting individual rats using a miniature microphone

Vocalizations are pivotal in mammalian communication, especially in humans. Rodents accordingly rely on ultrasonic vocalizations (USVs) that reflect their internal state as a primary channel during social interactions. However, attributing vocalizations to specific individuals remains challenging, impeding internal state assessment. Rats emit 50-kHz USVs to indicate positive states and intensify sniffing during alertness and social interactions. Here, we present a method involving a miniature microphone attached to the rat nasal cavity that allows to capture both male and female individual rat vocalizations and sniffing patterns during social interactions. We found that while the emission of 50-kHz USVs increases during close interactions, these signals lack specific behavioral associations. Moreover, a previously unreported low-frequency vocalization type marking rat social interactions was uncovered. Finally, different dynamics of sniffing and vocalization activities point to distinct underlying internal states. Thus, our method facilitates the exploration of internal states concurrent with social behaviors.

Evidence for vocal signatures and voice-prints in a wild parrot.

In humans, identity is partly encoded in a voice-print that is carried across multiple vocalizations. Other species also signal vocal identity in calls, such as shown in the contact call of parrots. However, it remains unclear to what extent other call types in parrots are individually distinct, and whether there is an analogous voice-print across calls. Here we test if an individual signature is present in other call types, how stable this signature is, and if parrots exhibit voice-prints across call types. We recorded 5599 vocalizations from 229 individually marked monk parakeets (Myiopsitta monachus) over a 2-year period in Barcelona, Spain. We examined five distinct call types, finding evidence for an individual signature in three. We further show that in the contact call, while birds are individually distinct, the calls are more variable than previously assumed, changing over short time scales (seconds to minutes). Finally, we provide evidence for voice-prints across multiple call types, with a discriminant function being able to predict caller identity across call types. This suggests that monk parakeets may be able to use vocal cues to recognize conspecifics, even across vocalization types and without necessarily needing active vocal signatures of identity.

Parental verbal responsiveness to infant vocalizations from 9 to 14 months of age

Parental communication styles and language usage may undergo changes during the course of child development, exhibiting variations across language backgrounds and cultures. This study aimed to explore how infant-parent interactions within Korean-speaking environments evolve over time through meticulous coding of day-long home audio recordings. The study examined whether the ratio and types of parental verbal responses vary based on infants’ age and vocalization types. A total of 16 infants and their parents participated in all-day home recordings using the Language Environment Analysis (LENA) system. The recordings were conducted in two rounds per family when the infants were aged 9–11 months and 12–14 months, with a three-month gap between each round. The frequency and types of infant vocalizations were analyzed and the contingency and types of parental verbal responsiveness were determined based on semantic and phonetic connection, as well as temporal appropriateness. The results showed that parents did not verbally respond to approximately 50 % of the infant vocalizations in the natural home environment. However, parents’ lack of verbal responses decreased significantly, and their contingent responses increased significantly with infant age. Parents were also not selectively responsive to infants’ canonical vocalizations over non-canonical vocalizations. Nevertheless, parents demonstrated a higher frequency of responses that were not only linguistically meaningful but also socially appropriate and contextually relevant to infants’ vocalizations as infants developed, which may play a significant role in scaffolding speech and language development.

Threat vocalisations are acoustically similar between humans (Homo sapiens) and chimpanzees (Pan troglodytes)

ABSTRACT In behavioural contexts like fighting, eating, and playing, acoustically distinctive vocalisations are produced across many mammalian species. Such expressions may be conserved in evolution, pointing to the possibility of acoustic regularities in the vocalisations of phylogenetically related species. Here, we test this hypothesis by comparing the degree of acoustic similarity between human and chimpanzee vocalisations produced in 10 similar behavioural contexts. We use two complementary analysis methods: Pairwise acoustic distance measures and acoustic separability metrics based on unsupervised learning algorithms. Cross-context analysis revealed that acoustic features of vocalisations produced when threatening another individual were distinct from other types of vocalisations and highly similar across species. Using a multimethod approach, these findings demonstrate that human vocalisations produced when threatening another person are acoustically similar to chimpanzee vocalisations in the same situation as compared to other types of vocalisations, likely reflecting a phylogenetically ancient vocal signalling system.

Relative divergence of mobbing calls and songs structures in passerine birds

Divergence and convergence in acoustic signals may be driven by different processes. Uncertainty about which of these processes best explains the patterns of acoustic variation between species has fuelled a long-standing debate in evolutionary biology. In particular, the features of acoustic signals are expected to vary according to their functional support. To compare the relative divergence of vocalizations according to their function, we examine two types of signals within 23 bird species, whilst controlling for species size, phylogeny and within-species variation: (i) mobbing calls emitted to recruit both conspecifics and heterospecifics against a predator and (ii) territorial songs, a signal playing an important role in sexual and social communication. We found that divergence of acoustic features varies according to their function (mobbing calls vs. territorial songs). Furthermore, species size influences spectral features, while phylogeny explained acoustic variation in only one of the variables measured: note richness. Finally, our results reveal that the acoustic characteristics can vary greatly within species, regardless of the vocalization type, indicating that such variations cannot be ignored when performing comparative analyses across species.

Ultrasonic vocalization phenotypes in the Ts65Dn and Dp(16)1Yey mouse models of Down syndrome

Down syndrome (DS) is a developmental disorder associated with a high incidence of challenges in vocal communication. DS can involve medical co-morbidities and structural social factors that may impact communication outcomes, which can present difficulties for the study of vocal communication challenges. Mouse models of DS may be used to study vocal communication differences associated with this syndrome and allow for greater control and consistency of environmental factors. Prior work has demonstrated differences in ultrasonic vocalization (USV) of the Ts65Dn mouse model of DS at a young adult age, however it is not known how USV characteristics are manifested at mature ages. Given that the aging process and age-related co-morbidities may also impact communication in DS, addressing this gap in knowledge may be of value for efforts to understand communication difficulties in DS across the lifespan. The current study hypothesized that the Ts65Dn and Dp(16)1Yey mouse models of DS would demonstrate differences in multiple measures of USV communication at a mature adult age of 5 months.Methods: Ts65Dn mice (n = 16) and euploid controls (n = 19), as well as Dp(16)1Yey mice (n = 20) and wild-type controls (n = 22), were evaluated at 5 months of age for USV production using a mating paradigm. Video footage of USV sessions were analyzed to quantify social behaviors of male mice during USV testing sessions. USV recordings were analyzed using Deepsqueak software to identify 10 vocalization types, which were quantified for 11 acoustic measures. Results: Ts65Dn, but not Dp(16)1Yey, showed significantly lower proportions of USVs classified as Step Up, Short, and Frequency Steps, and significantly higher proportions of USVs classified as Inverted U, than euploid controls. Both Ts65Dn and Dp(16)1Yey groups had significantly greater values for power and tonality for USVs than respective control groups. While Ts65Dn showed lower frequencies than controls, Dp(16)1Yey showed higher frequencies than controls. Finally, Ts65Dn showed reductions in a measure of complexity for some call types. No significant differences between genotype groups were identified in analysis of behaviors during testing sessions.Conclusion: While both Ts65Dn and Dp(16)1Yey show significant differences in USV measures at 5 months of age, of the two models, Ts65Dn shows a relatively greater numbers of differences. Characterization of communication phenotypes in mouse models of DS may be helpful in laying the foundation for future translational advances in the area of communication difficulties associated with DS.

Representation of conspecific vocalizations in amygdala of awake marmosets.

Human speech and animal vocalizations are important for social communication and animal survival. Neurons in the auditory pathway are responsive to a range of sounds, from elementary sound features to complex acoustic sounds. For social communication, responses to distinct patterns of vocalization are usually highly specific to an individual conspecific call, in some species. This includes the specificity of sound patterns and embedded biological information. We conducted single-unit recordings in the amygdala of awake marmosets and presented calls used in marmoset communication, calls of other species and calls from specific marmoset individuals. We found that some neurons (47/262) in the amygdala distinguished 'Phee' calls from vocalizations of other animals and other types of marmoset vocalizations. Interestingly, a subset of Phee-responsive neurons (22/47) also exhibited selectivity to one out of the three Phees from two different 'caller' marmosets. Our findings suggest that, while it has traditionally been considered the key structure in the limbic system, the amygdala also represents a critical stage of socially relevant auditory perceptual processing.

Deep audio embeddings for vocalisation clustering.

The study of non-human animals' communication systems generally relies on the transcription of vocal sequences using a finite set of discrete units. This set is referred to as a vocal repertoire, which is specific to a species or a sub-group of a species. When conducted by human experts, the formal description of vocal repertoires can be laborious and/or biased. This motivates computerised assistance for this procedure, for which machine learning algorithms represent a good opportunity. Unsupervised clustering algorithms are suited for grouping close points together, provided a relevant representation. This paper therefore studies a new method for encoding vocalisations, allowing for automatic clustering to alleviate vocal repertoire characterisation. Borrowing from deep representation learning, we use a convolutional auto-encoder network to learn an abstract representation of vocalisations. We report on the quality of the learnt representation, as well as of state of the art methods, by quantifying their agreement with expert labelled vocalisation types from 8 datasets of other studies across 6 species (birds and marine mammals). With this benchmark, we demonstrate that using auto-encoders improves the relevance of vocalisation representation which serves repertoire characterisation using a very limited number of settings. We also publish a Python package for the bioacoustic community to train their own vocalisation auto-encoders or use a pretrained encoder to browse vocal repertoires and ease unit wise annotation.

Seabird’s cry: repertoire and vocal expression of contextual valence in the little auk (Alle alle)

Many seabird species breed in colonies counting up to hundreds of thousands of individuals. Life in such crowded colonies might require special coding–decoding systems to reliably convey information through acoustic cues. This can include, for example, developing complex vocal repertoires and adjusting the properties of their vocal signals to communicate behavioural contexts, and thus regulate social interactions with their conspecifics. We studied vocalisations produced by the little auk (Alle alle)—a highly vocal, colonial seabird—over mating and incubation periods on the SW coast of Svalbard. Using passive acoustic recordings registered in a breeding colony, we extracted eight vocalisation types: single call, clucking, classic call, low trill, short call, short-trill, terror, and handling vocalisation. Calls were grouped by production context (based on the typically associated behaviour), to which a valence (positive vs negative) was later attributed, when possible, according to fitness threats, i.e. predator or human presence (negative) and promoters, i.e. interaction with a partner (positive). The effect of the putative valence on eight selected frequency and duration variables was then investigated. The putative contextual valence significantly affected the acoustic properties of the calls. Calls assigned positive valence had higher fundamental frequency and spectral centre of gravity as well as shorter sound duration than these assigned negative valence. These results indicate that the little auk’s vocal communication system may facilitate expression of complex behavioural contexts, and seems to include vocal plasticity within vocalisation types—however, more data are necessary to better understand this effect and possible interplays of other factors.