Neurocognitive processing efficiency for discriminating human non-alarm rather than alarm scream calls

Across many species, scream calls signal the affective significance of events to other agents. Scream calls were often thought to be of generic alarming and fearful nature, to signal potential threats, with instantaneous, involuntary, and accurate recognition by perceivers. However, scream calls are more diverse in their affective signaling nature than being limited to fearfully alarming a threat, and thus the broader sociobiological relevance of various scream types is unclear. Here we used 4 different psychoacoustic, perceptual decision-making, and neuroimaging experiments in humans to demonstrate the existence of at least 6 psychoacoustically distinctive types of scream calls of both alarming and non-alarming nature, rather than there being only screams caused by fear or aggression. Second, based on perceptual and processing sensitivity measures for decision-making during scream recognition, we found that alarm screams (with some exceptions) were overall discriminated the worst, were responded to the slowest, and were associated with a lower perceptual sensitivity for their recognition compared with non-alarm screams. Third, the neural processing of alarm compared with non-alarm screams during an implicit processing task elicited only minimal neural signal and connectivity in perceivers, contrary to the frequent assumption of a threat processing bias of the primate neural system. These findings show that scream calls are more diverse in their signaling and communicative nature in humans than previously assumed, and, in contrast to a commonly observed threat processing bias in perceptual discriminations and neural processes, we found that especially non-alarm screams, and positive screams in particular, seem to have higher efficiency in speeded discriminations and the implicit neural processing of various scream types in humans.

Human screams are more diverse in their communicative nature than those of other species, and are not limited to alarm signals of threat. This study shows that surprisingly, non-alarming screams, and positive screams in particular, have higher efficiency of their cognitive and neural processing than alarm screams.     

Rhesus monkey (Macaca mulatta) screams: Representational signalling in the recruitment of agonistic aid

Free-ranging rhesus monkeys on Cayo Santiago (Puerto Rico) give five acoustically distinct scream vocalizations during agonistic encounters. These calls are thought to be an important mechanism in the recruitment of support from allies against opponents. Alliance formation during agonistic encounters is known to vary with the dominance rank and matrilineal relatedness of opponents, as well as with the severity of aggression. In contrast to previous interpretations of screams as graded signals reflecting the level of arousal of the caller, we found these calls to be much more discrete, with each of the five acoustic types significantly associated with a particular class of opponent and level of physical aggression. We performed a series of field experiments in which tape-recorded screams of immature rhesus monkeys were played to their mothers in the absence of any other information. The results suggest that the information necessary for differential responses is conveyed by the scream vocalizations themselves. We conclude that screams are representational signals that refer to external objects and events and function in the system of agonistic alliance formation.     

Discrete or graded variation within rhesus monkey screams? Psychophysical experiments on classification

Gouzoules et al. (1984, Animal Behaviour,32, 182-193) presented evidence that semifree-ranging rhesus monkeys, Macaca mulatta, produce acoustically distinctive classes of scream vocalizations that carry different functional messages. To determine the perceptual validity of these vocal classes, we conducted psychophysical experiments on captive rhesus monkeys. We trained two monkeys to maintain contact with a metal response cylinder during presentation of nontarget stimuli, and to release the cylinder to report detection of target stimuli. For one subject, tonal screams served as nontarget stimuli and arched screams served as targets. These conditions were reversed for a second subject. Once natural exemplars were correctly discriminated, both subjects correctly generalized to synthetic targets. Variability in responses to nontarget stimuli, however, suggested that scream categories were not well defined following training. This result suggests that rhesus monkeys do not perceive categorical distinctions between arched and tonal screams, at least under the testing conditions implemented. Rather, our results provide evidence for a graded category. To explore which acoustic features are most important for classifying novel exemplars as tonal or arched screams, we ran several follow-up experiments with novel scream exemplars. Generalization trials suggested that variation in rate of frequency change, maximum frequency of the fundamental and harmonic structure may be important to the discrimination of screams.     

BIOACOUSTIC PUBLICATIONS, 1989

ABSTRACT

Coruros Spalacopus cyanus, social fossorial rodents from Chile, use a complex acoustic repertoire with eleven different true vocalisations and one mechanical sound in various behavioural contexts. The complex of contact calls is particularly well differentiated. Juvenile coruros produced six true vocalisations of which four were structurally identical to adult calls. One vocalisation had components of two adult sounds and one occurred only in juvenile animals. Certain calls from the adult repertoire were lacking. The frequencies of sounds of juveniles were considerably higher than those of adults, with many sounds reaching the ultrasonic range. Nevertheless, pure ultrasonic sounds were not recorded.

The frequencies of the analysed sounds of coruros extended from 0.17 to 20.33 kHz with dominant frequency components between 0.17 and 10 kHz. The acoustic properties of calls are suitable for transmission above and below ground, thus providing further indirect evidence that coruros are not strictly confined to an underground way of life. Indeed, the great variability of frequency ranges, with lower frequencies always being included, reflects a specialisation for communication in variable acoustic environments.

The most distinctive and unique vocalisation of coruros is the long duration musical trilling (lasting up to two minutes), which is a long-distance call emitted in alarm and arousal contexts. Recordings of this call from natural burrows in the field in Chile showed similar structural features to vocalisations from captive colonies in the laboratory.

Our findings provide a further example of matching physical properties of vocalisations to the acoustic conditions of the habitat. However, vocalisations in subterranean rodents consist almost exclusively of short-distance calls, the trilling of coruros being the notable exception. Since the selective pressure of the acoustic environment upon the evolution of short-distance vocalisations is probably minimal, we suggest that during their evolution, subterranean mammals have matched their vocalisations primarily to their hearing range and not directly to the acoustics underground. Hearing probably has been the primary target of natural selection, serving not only for communication but also for detection of predators (and, in carnivores, of prey).     

SILIC: A cross database framework for automatically extracting robust biodiversity information from soundscape recordings based on object detection and a tiny training dataset

• 1.Passive acoustic monitoring (PAM) offers many advantages comparing with other survey methods and gains an increasing use in terrestrial ecology, but the massive effort needed to extract species information from a large number of recordings limits its application. The convolutional neural network (CNN) has been demonstrated with its high performance and effectiveness in identifying sound sources automatically. However, requiring a large amount of training data still constitutes a challenge.
• 2.Object detection is used to detect multiple objects in photos or videos and is effective at detecting small objects in a complex context, such as animal sounds in a spectrogram and shows the opportunity to build a good performance model with a small training dataset. Therefore, we developed the Sound Identification and Labeling Intelligence for Creatures (SILIC), which integrates online animal sound databases, PAM databases and an object detection-based model, for extracting information on the sounds of multiple species from complex soundscape recordings.
• 3.We used the sounds of six owl species in Taiwan to demonstrate the effectiveness, efficiency and application potential of the SILIC framework. Using only 786 sound labels in 133 recordings, our model successfully identified the species' sounds from the recordings collected at five PAM stations, with a macro-average AUC of 0.89 and a mAP of 0.83. The model also provided the time and frequency information, such as the duration and bandwidth, of the sounds.
• 4.To our best knowledge, this is the first time that the object detection algorithm has been used to identify sounds of multiple wildlife species. With an online sound-labeling platform embedded and a novel data preprocessing approach (i.e., rainbow mapping) applied, the SILIC shows its good performance and high efficiency in identifying wildlife sounds and extracting robust species, time and frequency information from a massive amount of soundscape recordings based on a tiny training dataset acquired from existing animal sound databases. The SILIC can help expand the application of PAM as a tool to evaluate the state of and detect the change in biodiversity by, for example, providing high temporal resolution and continuous information on species presence across a monitoring network.

     

Automatic recording of individual oestrus vocalisation in group-housed dairy cattle: development of a cattle call monitor

Oestrus detection remains a problem in the dairy cattle industry. Therefore, automatic detection systems have been developed to detect specific behavioural changes at oestrus. Vocal behaviour has not been considered in such automatic oestrus detection systems in cattle, though the vocalisation rate is known to increase during oestrus. The main challenge in using vocalisation to detect oestrus is correctly identifying the calling individual when animals are moving freely in large groups, as oestrus needs to be detected at an individual level. Therefore, we aimed to automate vocalisation recording and caller identification in group-housed dairy cows. This paper first presents the details of such a system and then presents the results of a pilot study validating its functionality, in which the automatic detection of calls from individual heifers was compared to video-based assessment of these calls by a trained human observer, a technique that has, until now, been considered the ‘gold standard’. We developed a collar-based cattle call monitor (CCM) with structure-borne and airborne sound microphones and a recording unit and developed a postprocessing algorithm to identify the caller by matching the information from both microphones. Five group-housed heifers, each in the perioestrus or oestrus period, were equipped with a CCM prototype for 5 days. The recorded audio data were subsequently analysed and compared with audiovisual recordings. Overall, 1404 vocalisations from the focus heifers and 721 vocalisations from group mates were obtained. Vocalisations during collar changes or malfunctions of the CCM were omitted from the evaluation. The results showed that the CCM had a sensitivity of 87% and a specificity of 94%. The negative and positive predictive values were 80% and 96%, respectively. These results show that the detection of individual vocalisations and the correct identification of callers are possible, even in freely moving group-housed cattle. The results are promising for the future use of vocalisation in automatic oestrus detection systems.     

Expression of Emotional Arousal in Two Different Piglet Call Types

Humans as well as many animal species reveal their emotional state in their voice. Vocal features show strikingly similar correlation patterns with emotional states across mammalian species, suggesting that the vocal expression of emotion follows highly conserved signalling rules. To fully understand the principles of emotional signalling in mammals it is, however, necessary to also account for any inconsistencies in the way that they are acoustically encoded. Here we investigate whether the expression of emotions differs between call types produced by the same species. We compare the acoustic structure of two common piglet calls—the scream (a distress call) and the grunt (a contact call)—across three levels of arousal in a negative situation. We find that while the central frequency of calls increases with arousal in both call types, the amplitude and tonal quality (harmonic-to-noise ratio) show contrasting patterns: as arousal increased, the intensity also increased in screams, but not in grunts, while the harmonicity increased in screams but decreased in grunts. Our results suggest that the expression of arousal depends on the function and acoustic specificity of the call type. The fact that more vocal features varied with arousal in scream calls than in grunts is consistent with the idea that distress calls have evolved to convey information about emotional arousal.     

Listening to speech in the presence of other sounds

Although most research on the perception of speech has been conducted with speech presented without any competing sounds, we almost always listen to speech against a background of other sounds which we are adept at ignoring. Nevertheless, such additional irrelevant sounds can cause severe problems for speech recognition algorithms and for the hard of hearing as well as posing a challenge to theories of speech perception. A variety of different problems are created by the presence of additional sound sources: detection of features that are partially masked, allocation of detected features to the appropriate sound sources and recognition of sounds on the basis of partial information. The separation of sounds is arousing substantial attention in psychoacoustics and in computer science. An effective solution to the problem of separating sounds would have important practical applications.     

基于MFCC和GMM的昆虫声音自动识别

昆虫的运动、 取食、 鸣叫都会发出声音, 这些声音存在种内相似性和种间差异性, 因此可用来识别昆虫的种类。基于昆虫声音的昆虫种类自动检测技术对协助农业和林业从业人员方便地识别昆虫种类非常有意义。本研究采用了语音识别领域里的声音参数化技术来实现昆虫的声音自动鉴别。声音样本经预处理后, 提取梅尔倒谱系数（Mel frequency cepstrum coefficient, MFCC）作为特征, 并用这些样本提取的MFCC特征集训练混合高斯模型（Gaussian mixture model, GMM）。最后用训练所得到的GMM对未知类别的昆虫声音样本进行分类。该方法在包含58种昆虫声音的样本库中进行了评估, 取得了较高的识别正确率（平均精度为98.95%）和较理想的时间性能。该测试结果证明了基于MFCC和GMM的语音参数化技术可以用来有效地识别昆虫种类。     

A PORTABLE DIGITAL SOUND EMITTER FOR INTERACTIVE PLAYBACK OF ANIMAL VOCALISATIONS

ABSTRACT

A new portable digital sound emitter (DSE) for normal and interactive playback of sound signals in the field and in the laboratory is described together with two examples of applications of the DSE in interactive field experiments. The DSE may be loaded with a broad spectrum of digitised sounds, e.g. artificial signals or natural animal vocalisations, and it is controlled via the keyboard of a portable PC running a dedicated program. This program, which can be tailored to support a vast number of different demands, enables the experimenter to start and stop the analog output from the DSE at any time and to choose freely between the available sounds and playback modes. The use of the DSE for interactive playback experiments therefore is only limited by the ability of the operator to perceive the vocalisations of the participating animal and to operate the keyboard of the PC.     

Courtship Sounds Advertise Species Identity and Male Quality in Sympatric Pomatoschistus spp. Gobies

Acoustic signals can encode crucial information about species identity and individual quality. We recorded and compared male courtship drum sounds of the sand goby Pomatoschistus minutus and the painted goby P. pictus and examined if they can function in species recognition within sympatric populations. We also examined which acoustic features are related to male quality and the factors that affect female courtship in the sand goby, to determine whether vocalisations potentially play a role in mate assessment. Drums produced by the painted goby showed significantly higher dominant frequencies, higher sound pulse repetition rates and longer intervals between sounds than those of the sand goby. In the sand goby, male quality was predicted by visual and acoustic courtship signals. Regression analyses showed that sound amplitude was a good predictor of male length, whereas the duration of nest behaviour and active calling rate (i.e. excluding silent periods) were good predictors of male condition factor and fat reserves respectively. In addition, the level of female courtship was predicted by male nest behaviour. The results suggest that the frequency and temporal patterns of sounds can encode species identity, whereas sound amplitude and calling activity reflects male size and fat reserves. Visual courtship duration (nest-related behaviour) also seems relevant to mate choice, since it reflects male condition and is related to female courtship. Our work suggests that acoustic communication can contribute to mate choice in the sand goby group, and invites further study.     

Detection of bronchial breathing caused by pneumonia.

The classic auscultation with stethoscope is the established clinical method for the detection of lung diseases. The interpretation of the sounds depends on the experience of the investigating physician. Therefore, a new computer-based method has been developed to classify breath sounds from digital lung sound recordings. Lung sounds of 11 patients with one-sided pneumonia and bronchial breathing were recorded on both the pneumonia side and on contralateral healthy side simultaneously using two microphones. The spectral power for the 300-600 Hz frequency band was computed for four respiratory cycles and normalized. For each breath, the ratio R between the time-segments (duration = 0.1 s) with the highest inspiratory and highest expiratory flow was calculated and averaged. We found significant differences in R between the pneumonia side (R = 1.4 +/- 1.3) and the healthy side (R = 0.5 +/- 0.5; p = 0.003 Wilcoxon-test) of lung. In 218 healthy volunteers we found R = 0.3 +/- 0.2 as a reference-value. The differences of ratio R (delta R) between the pneumonia side and the healthy side (delta R = 1.0 +/- 0.9) were significantly higher compared to follow-up studies after recovery (delta R = 0.0 +/- 0.1, p = 0.005 Wilcoxon-test). The computer based detection of bronchial breathing can be considered useful as part of a quantitative monitoring of patients at risk to develop pneumonia.     

Heart energy signature spectrogram for cardiovascular diagnosis

A new method and application is proposed to characterize intensity and pitch of human heart sounds and murmurs. Using recorded heart sounds from the library of one of the authors, a visual map of heart sound energy was established. Both normal and abnormal heart sound recordings were studied. Representation is based on Wigner-Ville joint time-frequency transformations. The proposed methodology separates acoustic contributions of cardiac events simultaneously in pitch, time and energy. The resolution accuracy is superior to any other existing spectrogram method. The characteristic energy signature of the innocent heart murmur in a child with the S3 sound is presented. It allows clear detection of S1, S2 and S3 sounds, S2 split, systolic murmur, and intensity of these components. The original signal, heart sound power change with time, time-averaged frequency, energy density spectra and instantaneous variations of power and frequency/pitch with time, are presented. These data allow full quantitative characterization of heart sounds and murmurs. High accuracy in both time and pitch resolution is demonstrated. Resulting visual images have self-referencing quality, whereby individual features and their changes become immediately obvious.     

Screening large audio datasets to determine the time and space distribution of Screaming Piha birds in a tropical forest

Acoustic monitoring has proved to be an efficient approach to monitor wildlife, notably in environments with limited visibility, such as tropical rainforests. Today, recording equipment allows acoustic data to be gathered in remote areas at wide spatial and temporal scales. The resulting datasets are large and the use of automated processing systems to extract relevant information can greatly facilitate their analysis. Here, we have developed signal processing techniques to reveal the spatio-temporal dynamics of an emblematic bird voice of the neotropical forest: the song of the Screaming Piha (Lipaugus vociferans). Using recordings made in a French Guianan lowland forest, with an array of 24 microphones in a three dimensional space, we implemented a detection system based on spectrogram cross-correlation to trace the vocalisations of L. vociferans. We tuned the detection system based on the percentage area under the Receiver Operating Characteristic curve, finding a maximum of 95.88%. To strictly minimise false positives, we set the operating point to have 34.9% true positives and 0% false positives. We detected a total of 12,735 songs attributed to the study bird during 25 study days. We found that spatial patterns of lower activity corresponded to a zone having smaller trees and more tree gaps — a known liana forest patch — suggesting that Screaming Piha birds tend to avoid non-mature primary forests. The sampling sites near the creeks had more detections than the sites further away, suggesting that the lek mating arenas might be distributed strategically to be near to a source of water. We also found a marked temporal pattern. The lek was active during the whole day, from sunrise to sunset, with two peaks of activity shifted by more than 2 h from the dawn and dusk chorus. The approach described here can be tested using other conspicuous and stereotyped sounds that occur within a heterogeneous and noisy background. To decipher the complex interacting sounds of the tropical forest, these focal studies on specific acoustic elements should be complemented with community or soundscape analysis, to demonstrate the human impact on the ecosystem and to provide guidelines for natural resource management.     

Do film soundtracks contain nonlinear analogues to influence emotion?

A variety of vertebrates produce nonlinear vocalizations when they are under duress. By their very nature, vocalizations containing nonlinearities may sound harsh and are somewhat unpredictable; observations that are consistent with them being particularly evocative to those hearing them. We tested the hypothesis that humans capitalize on this seemingly widespread vertebrate response by creating nonlinear analogues in film soundtracks to evoke particular emotions. We used lists of highly regarded films to generate a set of highly ranked action/adventure, dramatic, horror and war films. We then scored the presence of a variety of nonlinear analogues in these film soundtracks. Dramatic films suppressed noise of all types, contained more abrupt frequency transitions and musical sidebands, and fewer noisy screams than expected. Horror films suppressed abrupt frequency transitions and musical sidebands, but had more non-musical sidebands, and noisy screams than expected. Adventure films had more male screams than expected. Together, our results suggest that film-makers manipulate sounds to create nonlinear analogues in order to manipulate our emotional responses.     

Transmission characteristics of primate vocalizations: implications for acoustic analyses

Acoustic analyses have become a staple method in field studies of animal vocal communication, with nearly all investigations using computer-based approaches to extract specific features from sounds. Various algorithms can be used to extract acoustic variables that may then be related to variables such as individual identity, context or reproductive state. Habitat structure and recording conditions, however, have strong effects on the acoustic structure of sound signals. The purpose of this study was to identify which acoustic parameters reliably describe features of propagated sounds. We conducted broadcast experiments and examined the influence of habitat type, transmission height, and re-recording distance on the validity (deviation from the original sound) and reliability (variation within identical recording conditions) of acoustic features of different primate call types. Validity and reliability varied independently of each other in relation to habitat, transmission height, and re-recording distance, and depended strongly on the call type. The smallest deviations from the original sounds were obtained by a visually-controlled calculation of the fundamental frequency. Start- and end parameters of a sound were most susceptible to degradation in the environment. Because the recording conditions can have appreciable effects on acoustic parameters, it is advisable to validate the extraction method of acoustic variables from recordings over longer distances before using them in acoustic analyses.     

Vocal Mimicry of Alarm‐Associated Sounds by a Drongo Elicits Flee and Mobbing Responses from Other Species that Participate in Mixed‐Species Bird Flocks

A growing number of studies have shown that vocal mimicry appears to be adaptive for some bird species, although the exact function of this behaviour varies among species. Previous work has looked at the function of the vocal mimicry of non‐alarm sounds by the Greater Racket‐tailed Drongo (Dicurus paradiseus). But drongos also imitate sounds associated with danger, such as predators' vocalisations or the mobbing‐specific vocalisations of other prey species, raising the question of whether the function of mimicry can vary even within a species. In a playback experiment, we compared the effect on other species of different drongo vocalisations including: (1) predator mimicry, (2) mobbing mimicry, (3) drongo species‐specific alarms, (4) drongo species‐specific non‐alarms and (5) a control (barbet) sound. Both mobbing mimicry and drongo species‐specific alarms elicited flee responses from the most numerous species in the flocks, the Orange‐billed Babbler (Turdoides rufescens). Mobbing mimicry also elicited mobbing responses from the Orange‐billed Babbler and from another gregarious babbler, the Ashy‐headed Laughingthrush (Garrulax cinereifrons); when responses from both species were considered together, they were elicited at a significantly higher level by mobbing mimicry than by the barbet control, and a level that tended to be higher (0.07 < p < 0.10) than the response to drongo‐specific alarms. Predator mimicry elicited flee and mobbing responses at an intermediary level. Our results support the hypotheses that mobbing mimicry is a specific category of mimicry that helps attract the aid of heterospecifics during mobbing and that alarm mimicry can in some cases be beneficial to the caller.     

A lightweight CNN-based model for early warning in sow oestrus sound monitoring

The reproductive performance of sows is an important indicator for evaluating the economic efficiency and production level of pigs. In this paper, we design and propose a lightweight sow oestrus detection method based on acoustic data and deep convolutional neural network (CNN) algorithms by collecting and analysing short-frequency and long-frequency sow oestrus sounds. We use visual log-mel spectrograms, which can reflect three-dimensional information, as inputs to the network model to improve the overall recognition accuracy. The improved lightweight MobileNetV3_esnet model is used to identify oestrus and nonoestrus sounds and is compared with existing algorithms. The model outperforms the other algorithms, with 97.12% precision, 97.34% recall, 97.59% F1-score, and 97.52% accuracy; the model size is 5.94 MB. Compared with traditional oestrus monitoring methods, the proposed method can more accurately boost the vocal characteristics exhibited by sows in latent oestrus, thus providing an efficient and accurate approach for use in practical applications of oestrus monitoring and early warning systems on pig farms.