[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121693-en":3,"doc-seo-121693-105":31,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},121693,549758146520,"Patrick","https://ap-avatar.wpscdn.com/avatar/80002397d8c0411e94?_k=1775819394049821470",6,"Technology","Whombat - An Open-Source Annotation Tool for Machine Learning Development in Bioacoustics","Machine-learning-based analysis of bioacoustic recordings can scale biodiversity monitoring, but high-stakes conservation and research require a data-centric workflow built on carefully curated, representative training and evaluation data. Creating annotated sound datasets is difficult due to large recording collections with complex metadata, the need for flexible tools covering diverse vocalization profiles, and the shortage of expert annotators. Whombat addresses these needs with a browser-based interface for managing audio and annotations plus visualization, exploration, and iterative feedback with ML predictions.","A P P L ICAT IO NS  \narXiv :2308 . 12688v1 [ cs . SD] 24 Aug 2023  \nWhombat: An open-source annotation tool for machine learning development in bioacoustics  \nSantiago Martínez Balvanera1 | Oisin Mac Aodha2 | Matthew J. Weldy3,4 | Holly Pringle1 | Ella Browning1,5 | Kate E. Jones1  \n1 Centre for Biodiversity and Environment Research, Department of Genetics, Evolution and Environment, University College London, London, WC1E 6BT, United Kingdom  \n2School of Informatics, University of Edinburgh, Edinburgh, EH8 9AB, United Kingdom  \n3 Department of Forest Ecosystems and Society, College of Forestry, Oregon State University, Corvallis, OR 97331-5704, USA  \n4 Pacific Northwest Research Station, USDA Forest Service, Corvallis, USA  \n5 Bat Conservation Trust, Studio 15 Cloisters House, Cloisters Business Centre, 8 Battersea Park Road, London, SW8 4BG, United Kingdom  \nCorrespondence  \nSantiago Martínez Balvanera  \nEmail: [santiago.balvanera.20@ucl.ac.uk](santiago.balvanera.20@ucl.ac.uk)  \nFunding information  \nCONACYT, Grant/Award Number: 2020-000017-02EXTF-00334  \nAbstract  \n1. Automated analysis of bioacoustic recordings using machine learning (ML) methods has the potential to greatly scale biodiversity monitoring efforts. The use of ML for high-stakes applications, such as conservation and scientific research, demands a data-centric approach with a focus on selecting and utilizing carefully annotated and curated evaluation and training data that is relevant and representative. Creating annotated datasets of sound recordings presents a number of challenges, such as managing large collections of recordings with associated metadata, developing flexible annotation tools that can accommodate the diverse range of vocalization profiles of different organisms, and addressing the scarcity of expert annotators.  \n2. We present Whombat a user-friendly, browser-based interface for managing audio recordings and annotation projects, with several visualization, exploration, and annotation tools. It enables users to quickly annotate, review, and share annotations, as well as visualize and evaluate a set of machine learning predictions on a dataset. The tool facilitates an iterative workflow where user annotations and machine learning predictions feedback to enhance model performance and annotation quality.  \n3. We demonstrate the flexibility of Whombat by showcasing two distinct use cases: (1) an project aimed at enhancing automated UK bat call identification at the Bat Conservation Trust (BCT), and (2) a collaborative effort among the USDA Forest Service and Oregon State University researchers exploring bioacoustic applications and extending automated avian classification models in the Pacific Northwest, USA.  \n4. Whombat is a flexible tool that can effectively address the challenges of annotation for bioacoustic research. It can be used for individual and collaborative work, hosted on a shared server or accessed remotely, or  \nrun on a personal computer without the need for coding skills. The code is open-source, and we provide a user guide.  \nK EYW O R D S  \nAudio Annotation, Bioacoustics, Machine Learning, Software, Sound Event Detection, Visualization  \n1 | INTRODUCTION  \nRecent advancements in Machine Learning (ML) are revolutionizing our ability to analyse large datasets generated by passive acoustic recorders for ecologically relevant signals (Kitzes et al., 2021; Tuia et al., 2022) . Open-source Deep Learning models, such as BirdNET (Kahl et al., 2021) and NABat ML (Khalighifar et al., 2022), can be used to monitor birds and bats at scale across large regions. While considerable attention has been directed towards developing sophisticated ML systems, it is crucial to acknowledge the pivotal role of data and data work in establishing reliable ML implementations (Nithya Sambasivan etal., 2021) . In line with this, the data-centric approach has gained increasing relevance (Jarrahi, Mohammad Hossein et al., 2022), emphasizing the collection, curation, and ","cbCaicxnefuZED6P","https://ap.wps.com/l/cbCaicxnefuZED6P","pdf",2970098,2,1,17,"English","en",105,"# Introduction\n## Data-centric machine learning for bioacoustics\n## Annotation as an iterative feedback workflow\n## Challenges in bioacoustic dataset creation\n## Purpose and overview of Whombat","[{\"question\":\"Why is a data-centric approach important for machine learning in bioacoustics?\",\"answer\":\"Reliable ML in conservation and scientific research depends on selecting and using carefully annotated and representative training and evaluation data to ensure trustworthy model performance.\"},{\"question\":\"What challenges make bioacoustic annotation difficult?\",\"answer\":\"Key challenges include managing large recording collections with metadata, building flexible tools for varied vocalization profiles across organisms, and the limited availability of expert annotators.\"},{\"question\":\"How does Whombat support the annotation and model-improvement workflow?\",\"answer\":\"Whombat provides a browser-based interface to manage audio recordings and annotation projects, enabling users to annotate, review, share, and visualize ML predictions so that user annotations and predictions feed back iteratively to improve model performance and annotation quality.\"}]","Whombat - An Open-Source Annotation Tool for Machine Learning Development in Bioacoustics | PDF",1785806298,43,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":29},"whombat-an-open-source-annotation-tool-for-machine-learning-development-in-bioacoustics","",{"@graph":37,"@context":85},[38,54,68],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,48,51],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":20},"https://docshare.wps.com/document/","Document",{"item":49,"name":12,"@type":44,"position":50},"https://docshare.wps.com/document/technology/",3,{"item":52,"name":13,"@type":44,"position":53},"https://docshare.wps.com/document/whombat-an-open-source-annotation-tool-for-machine-learning-development-in-bioacoustics/121693/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is a data-centric approach important for machine learning in bioacoustics?","Question",{"text":75,"@type":76},"Reliable ML in conservation and scientific research depends on selecting and using carefully annotated and representative training and evaluation data to ensure trustworthy model performance.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What challenges make bioacoustic annotation difficult?",{"text":80,"@type":76},"Key challenges include managing large recording collections with metadata, building flexible tools for varied vocalization profiles across organisms, and the limited availability of expert annotators.",{"name":82,"@type":73,"acceptedAnswer":83},"How does Whombat support the annotation and model-improvement workflow?",{"text":84,"@type":76},"Whombat provides a browser-based interface to manage audio recordings and annotation projects, enabling users to annotate, review, share, and visualize ML predictions so that user annotations and predictions feed back iteratively to improve model performance and annotation quality.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,113,118,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":47,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":47,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":111,"slug":112},50,"technology",{"id":114,"doc_module":4,"doc_module_name":47,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":119,"doc_module":4,"doc_module_name":47,"category_name":120,"show_sort_weight":121,"slug":122},8,"Research & Report",30,"research-report",{"id":124,"doc_module":4,"doc_module_name":47,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":47,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":47,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":47,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]