Decoding vocal indicators of stress in laying hens: A CNN-MFCC deep learning framework
Bibliographic record
Abstract
Artificial intelligence is revolutionizing our capacity to interpret and respond to animal emotional states. This study leverages advanced Convolutional Neural Networks (CNNs) combined with Mel Frequency Cepstral Coefficients (MFCCs) to decode intricate vocalization patterns in laying hens experiencing acute environmental stress. Controlled exposure to realistic auditory stimuli (dog barking) and visual stimuli (umbrella opening) across different developmental stages enabled a critical comparative evaluation of vocal stress responses within a commercial-like experimental setup. Over five weeks, audio data were systematically captured from control and treatment groups, providing insights into vocal behaviors before and after stress induction. Remarkably, younger hens demonstrated significantly elevated vocal activity and more pronounced spectral shifts when stressed, underscoring age-dependent variations in emotional reactivity and coping mechanisms. The CNN model attained a remarkable 94% classification accuracy, reliably discriminating stressor types, age categories, and exposure conditions based solely on MFCC-derived acoustic signatures. Analysis further revealed that lower-order MFCC features are acutely responsive to stress-induced vocal dynamics, whereas higher-order coefficients remained relatively constant, signifying subtle emotional states. These compelling findings position vocalizations as powerful, non-invasive biomarkers of welfare status in poultry, supporting real-time, AI-driven monitoring solutions. By facilitating early, precise detection of distress signals, this innovative approach holds substantial promise for enhancing welfare standards and management decisions in livestock production. Ultimately, this study presents a robust, scalable methodology poised to advance digital agriculture broadly, turning previously silent animal expressions into essential indicators of their wellbeing and transforming farm animal management into a more ethically responsive practice.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".