Acoustic sensors for automated detection of cow vocalization duration and type

Shorten, Paul R.; Hunter, Laura B.

doi:10.1016/j.compag.2023.107760

Cited by 9 publications

(10 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As shown in Table 2 , both the explainable and DL models were able to accurately classify between the low- and high-frequency calls, with 87.2 and 89.4% accuracy, respectively. This outcome slightly outperforms (2%, 4.4%) the current state-of-the-art model ( 33 ), which used a smaller dataset of n = 10 individuals. Notably, the differences between the models’ performances between the training and testing cohorts was around 2.5% toward the training cohort, compared to the state-of-the-art, which reports a 14.2% difference.…”

Section: Discussionmentioning

confidence: 73%

“…Methods of studying animal vocal communication are becoming increasingly automated, with a growing body of research validating the use of both hardware and software that are capable of automatically collecting and processing bioacoustics data [reviewed by Mcloughlin et al ( 18 )]. In this vein, Shorten and Hunter ( 33 ) found significant variability in cattle vocalization parameters, and suggested that such traits can be monitored using animal-attached acoustic sensors in order to provide information on the welfare and emotional state of the animal. Therefore, automated vocalization monitoring could prove to be a useful tool in precision livestock farming ( 18 , 34 , 35 ), especially as dairy farming systems become increasingly automated with wide-scale use of milking and feeding robots, all this having the potential to dynamically adjust the management practices while the number of animals per farm unit tends to increase.…”

Section: Introductionmentioning

confidence: 99%

“…Machine learning techniques are therefore increasingly applied in the study of cattle vocalizations. Some tasks that have been addressed to date include the classification of high vs. low frequency calls ( 33 ), ingestive behavior ( 35 ), and categorization of calls such as oestrus and coughs ( 34 ).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

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

Gavojdian,

Mincu,

Lazebnik

et al. 2024

Front. Vet. Sci.

View full text Add to dashboard Cite

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.

show abstract

Section: Discussionmentioning

confidence: 73%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

Gavojdian,

Mincu,

Lazebnik

et al. 2024

Front. Vet. Sci.

View full text Add to dashboard Cite

show abstract

“…As shown in Table 2, both the explainable and DL models were able to accurately classify between the low-and highfrequency calls, with 87.2% and 89.4% accuracy, respectively. This outcome slightly outperforms (2%, 4,4%) the current state-of-the-art model 33 , which used a smaller dataset of n=10 individuals. Notably, the differences between the models' performances between the training and testing cohorts was around 2.5% towards the training cohort, compared to the stateof-the-art which reports a 14.2% difference.…”

Section: Discussionmentioning

confidence: 75%

“…Machine learning techniques are therefore increasingly applied in the study of cattle vocalizations. Some tasks that have been addressed to date include classification of high vs. low frequency calls 33 , ingestive behaviour 35 , and categorization of calls such as oestrus and coughs 34 .…”

Section: Introductionmentioning

confidence: 99%

BovineTalk: Machine Learning for Vocalization Analysis of Dairy Cattle under Negative Affective States

Gavojdian

Lazebnik

Mincu

et al. 2023

Preprint

View full text Add to dashboard Cite

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 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.

show abstract

Acoustic sensors to detect the rate of cow vocalization in a complex farm environment

Shorten,

Hunter

2024

Applied Animal Behaviour Science

View full text Add to dashboard Cite

Acoustic sensors for automated detection of cow vocalization duration and type

Cited by 9 publications

References 30 publications

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

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

BovineTalk: Machine Learning for Vocalization Analysis of Dairy Cattle under Negative Affective States

Acoustic sensors to detect the rate of cow vocalization in a complex farm environment

Contact Info

Product

Resources

About