Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2204.00331

Using segment-based features of jaw movements to recognize foraging activities in grazing cattle

José O. Chelotti·Sebastián R. Vanrell·Luciano S. Martinez-Rau·Julio R. Galli·Santiago A. Utsumi·Alejandra M. Planisich·Suyai A. Almirón·Diego H. Milone·Leonardo L. Giovanini·H. Leonardo Rufiner
View original source ↗Read source document ↗Export metadata ↓

Abstract

Precision livestock farming optimizes livestock production through the use of sensor information and communication technologies to support decision making, proactively and near real-time. Among available technologies to monitor foraging behavior, the acoustic method has been highly reliable and repeatable, but can be subject to further computational improvements to increase precision and specificity of recognition of foraging activities. In this study, an algorithm called Jaw Movement segment-based Foraging Activity Recognizer (JMFAR) is proposed. The method is based on the computation and analysis of temporal, statistical and spectral features of jaw movement sounds for detection of rumination and grazing bouts. They are called JM-segment features because they are extracted from a sound segment and expect to capture JM information of the whole segment rather than individual JMs. Two variants of the method are proposed and tested: (i) the temporal and statistical features only JMFAR-ns; and (ii) a feature selection process (JMFAR-sel). The JMFAR was tested on signals registered in a free grazing environment, achieving an average weighted F1-score of 93%. Then, it was compared with a state-of-the-art algorithm, showing improved performance for estimation of grazing bouts (+19%). The JMFAR-ns variant reduced the computational cost by 25.4%, but achieved a slightly lower performance than the JMFAR. The good performance and low computational cost of JMFAR-ns supports the feasibility of using this algorithm variant for real-time implementation in low-cost embedded systems. The method presented within this publication is protected by a pending patent application: AR P20220100910.

Explore related subjects

cs.SDeess.AS

Keep this discovery

BibTeXRIS

José O. Chelotti, Sebastián R. Vanrell, Luciano S. Martinez-Rau, Julio R. Galli, Santiago A. Utsumi, Alejandra M. Planisich, Suyai A. Almirón, Diego H. Milone, Leonardo L. Giovanini, H. Leonardo Rufiner. 2022-04-01. Using segment-based features of jaw movements to recognize foraging activities in grazing cattle. https://doi.org/10.1016/j.biosystemseng.2023.03.014

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Xiaomi-CocktailASR-1 Technical Report

Recently, large language model (LLM) based ASR models have achieved significant progress, yet they generally lack support for multi-speaker scenarios, where the cocktail party problem remains a critical bottleneck for further advancing ASR. Existing TS-ASR methods, including end-to-end architectures with speaker embeddings and latest LLM-based explorations suffer from degraded single-speaker performance and the inability to reject when the target speaker is absent. In this paper, we propose Xiaomi-CocktailASR-1, an LLM-based end-to-end TS-ASR architecture. By utilizing reference speech as voiceprint prompts, it directly transcribes the target speaker's speech without requiring speech separation. Xiaomi-CocktailASR-1 maintains competitive performance in single-speaker scenarios, comparable to mainstream ASR models. It also features a negative sample rejection capability, outputting empty text when the target speaker is absent from the mixed speech. Additionally, Xiaomi-CocktailASR-1 supports a Chain-of-Thought (CoT) reasoning mode to provide explicit reasoning steps. Extensive experiments on various synthetic and real-world multispeaker benchmarks demonstrate that Xiaomi-CocktailASR-1 achieves state-of-the-art performance, effectively addressing the cocktail party problem through a unified architecture that balances multispeaker and single-speaker recognition accuracy, along with rejection capability.

cs.SD↗
arXiv2026

EConv-TasNet: Efficient Conv-TasNet for Effective Speech Separation

Conv-TasNet has served as a strong baseline for time-domain speech separation, and many studies have extended it with advanced architectures such as dual-path networks, U-Nets, and attention mechanisms. However, these methods often introduce high computational cost and complexity, limiting their deployment in resource-constrained scenarios. To address this issue, we propose eConv-TasNet, an efficient variant of Conv-TasNet that improves both effectiveness and efficiency without relying on resource-intensive modules. The proposed model consists of a group-wise early-splitting (GES) module and a multi-group feature aggregation (MGFA) module. GES generates discriminative speaker embeddings at intermediate stages, while MGFA progressively aggregates these group-level representations for refined mask estimation. Experimental results show that eConv-TasNet reduces model size by 22.4%, accelerates inference by 18.9%, and improves SI-SNRi by 14.0%-28.0% across three public benchmarks. Moreover, it achieves competitive performance compared with state-of-the-art methods while requiring significantly fewer parameters and lower inference cost. These results demonstrate a favorable efficiency-effectiveness trade-off for edge deployment.

cs.SD↗