{"paper":{"title":"BioSEN: A Bio-acoustic Signal Enhancement Network for Animal Vocalizations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"BioSEN adapts speech enhancement methods into a lighter network that cleans animal vocalization recordings as well as or better than existing models.","cross_cats":["cs.LG","q-bio.NC"],"primary_cat":"cs.SD","authors_text":"Hisako Nomura, Linh Thi Hoai Nguyen, Ngamta Thamwattana, Tianyu Song, Ton Viet Ta","submitted_at":"2026-05-02T00:19:24Z","abstract_excerpt":"Most work in audio enhancement targets human speech, while bioacoustics is less studied due to noisy recordings and the distinct traits of animal sounds. To fill this gap, we adapt speech enhancement methods and build BioSEN, a model made for bioacoustic signals. BioSEN has three modules: a multi-scale dual-axis attention unit for time-frequency feature extraction, a bio-harmonic multi-scale enhancement unit for capturing harmonic structures, and an\n  energy-adaptive gating connection unit that uses frequency weights to keep vocalizations from being removed as noise. Tests on three bioacoustic"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Tests on three bioacoustic datasets show that BioSEN matches or exceeds state-of-the-art speech enhancement models while using far less computation.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That adaptations of speech enhancement methods with the described modules will generalize across diverse animal species and recording conditions without requiring extensive species-specific retraining or validation.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"BioSEN enhances bioacoustic signals with specialized modules and matches or exceeds speech enhancement models on animal datasets while using less computation.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"BioSEN adapts speech enhancement methods into a lighter network that cleans animal vocalization recordings as well as or better than existing models.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d92719616a02160290b38186105b899306386be582415bb9efe2a1ab5c483da9"},"source":{"id":"2605.12534","kind":"arxiv","version":2},"verdict":{"id":"d17633c4-029b-4d74-9865-af91e39eb8fa","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T07:33:21.943417Z","strongest_claim":"Tests on three bioacoustic datasets show that BioSEN matches or exceeds state-of-the-art speech enhancement models while using far less computation.","one_line_summary":"BioSEN enhances bioacoustic signals with specialized modules and matches or exceeds speech enhancement models on animal datasets while using less computation.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That adaptations of speech enhancement methods with the described modules will generalize across diverse animal species and recording conditions without requiring extensive species-specific retraining or validation.","pith_extraction_headline":"BioSEN adapts speech enhancement methods into a lighter network that cleans animal vocalization recordings as well as or better than existing models."},"references":{"count":23,"sample":[{"doi":"","year":2024,"title":"Kohlberg, A. B., Myers, C. R., Figueroa, L. L. (2024). Fro m buzzes to bytes: A sys- tematic review of automated bioacoustics models used to det ect, classify and monitor insects. J. Appl. Ecol. , 61(","work_id":"bc739ffd-ce72-4f4c-a1f2-5d86dcfc9ab9","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Navine, A. K., Camp, R. J., Weldy, M. J., Denton, T., Hart, P. J. (2024). Counting the chorus: A bioacoustic indicator of population density. Ecological Indicators, 169, 112930","work_id":"6770f918-b342-4111-ba88-242e88c2da79","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"H., Stowell, D., Briefer, E","work_id":"ca38ad9d-c965-4fb1-b0a1-7482520232d0","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Sharma, S., Sato, K., Gautam, B. P. (2023). A methodologi cal literature review of acoustic wildlife monitoring using artificial intelligenc e tools and techniques. Sustain- ability, 15(9), 7128","work_id":"2eafb687-9a4b-439a-8474-c962707aa70e","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Gajecki, T., Nogueira, W. (2025). Adversarial learning for end-to-end cochlear speech denoising using lightweight deep learning models. Proc. IEEE Int. Conf. Acoust., Speech, Signal Process. (ICASSP) ","work_id":"a8291412-c13c-4630-afd4-d8a26e9b95f1","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":23,"snapshot_sha256":"57afc62f5d80df8514fc7feab166b90d26b635c9022ed4a157e5d2f477ac43b5","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}