{"success":true,"data":{"author":"Imageomics Institute and ABC Global Center (Banerji N., Provost K. et al.)","bucketAccessType":"public","bucketName":"imageomics/upper-waiakea-PAM","bucketPrefix":"","category":"Speech & Audio","cloudProvider":"huggingface","createdAt":"2026-11-16T02:38:32.455Z","description":"About 1,623 hours of unlabelled passive acoustic monitoring (PAM) audio from 19 SongMeter Micro 2 recorders in the Upper Waiakea Forest Reserve on Hawaii Island, collected in January 2025 during the Imageomics Institute and ABC Global Center Experiential Introduction to AI and Ecology course. The soundscapes capture native Hawaiian forest birds in kipuka (forest islands surrounded by lava flows), for unsupervised audio analysis, running existing classifiers such as BirdNET or Perch, and relating song activity to habitat. Each recorder folder holds hourly WAV files named recorder_YYYYMMDD_HHMMSS and a summary text file; kipuka_metadata.csv gives recorder, SD card and point ids, deployment and retrieval dates and latitude and longitude; check.py summarises coverage. Measured: 1,698 files, 1,672 WAV, about 281 GB. No species labels; splits must be made by the user. Licensed CC BY 4.0 by the curators; cite the dataset.","displayTitle":"Upper Waiakea Forest Reserve Passive Acoustic Recordings of Hawaiian Forest Birds","id":"7b8f220c-f9c9-5bd5-9a4f-c0cccc98a3f5","license":"CC-BY-4.0","licenseEvidenceUrl":"https://huggingface.co/datasets/imageomics/upper-waiakea-PAM","objectCountEstimate":1698,"probeStatus":"ok","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"imageomics/upper-waiakea-PAM\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"imageomics/upper-waiakea-PAM\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=imageomics, license:cc-by-4.0, size_categories:1K<n<10K, downloads=419"],"format":"mixed","generatedAt":"2026-11-16T02:38:37.636Z","layout":{"example":"kipuka_metadata.csv","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"recorder_id,card_code,point_id,deployment_date,retrieval_date,latitude,longitude\n2MM01654,MSD-4525,-,,,,\n2MM01007,MSD-4524,k00,22-Jan-2025,28-Jan-2025,19.6749362,-155.3291983\n2MM01345,MSD-4534,k03,22-Jan-2025,28-Jan-2025,19.6702699,-155.3583000\n2MM01707,MSD-4532,k04,21-Jan-2025,28-Jan-2025,19.6678375,-155.3696367\n2MM00549,MSD-4536,k07,21-Jan-2025,28-Jan-2025,19.6547629,-155.3702900\n2MM01655,MSD-4516,k08,21-Jan-2025,28-Jan-2025,19.6516330,-155.3641660\n2MM01323,MSD-4531,k10,21-Jan-2025,28-Jan-2025,19.6480679,","encoding":"utf8","key":"kipuka_metadata.csv","truncatedAt":512}],"sampleObjects":["2MM00471/._2MM00471_Summary.txt","2MM00471/2MM00471_Summary.txt","2MM00471/Data/2MM00471_20250121_083657.wav","2MM00471/Data/2MM00471_20250122_132158.wav","2MM00471/Data/2MM00471_20250122_142103.wav"],"sidecars":[{"bytes":"---\nlicense: cc-by-4.0\nlanguage:\n- en\npretty_name: Upper Waiākea Forest Reserve Passive Recordings\ntask_categories:\n- audio-classification\nmodalities:\n- Audio\ntags:\n- biology\n- audio\n- bioacoustics\n- animals\n- CV\n- Hawaiian birds\n- kipuka\n- fragmented habitat\n- unlabeled\n- passive acoustic monitoring\nsize_categories: 1K<n<10K\ndescription: This is a dataset containing unlabelled, unprocessed passive acoustic recordings of Hawaiian birds in the Upper Waiākea Forst Reserve in Hawaii. This dataset is intended for use in unsupervised audio analysis methods, classification using existing models, and other machine learning and ecology research purposes. \n---\n\n<!--\n\nNOTE: Add more tags (your particular animal, type of model and use-case, etc.).\n\nAs with your GitHub Project repo, it is important to choose an appropriate license for your dataset. The default license is [CC0](https://creativecommons.org/publicdomain/zero/1.0/) (public domain dedication, see [Dryad's explanation of why to use CC0](https://blog.datadryad.org/2023/05/30/good-data-practices-removing-barriers-to-data-reuse-with-cc0-licensing/)). Alongside the appropriate stakeholders (eg., your PI, co-authors), select a license that is [Open Source Initiative](https://opensource.org/licenses) (OSI) compliant.\nFor more information on how to choose a license and why it matters, see [Choose A License](https://choosealicense.com) and [A Quick Guide to Software Licensing for the Scientist-Programmer](https://doi.org/10.1371/journal.pcbi.1002598) by A. Morin, et al.\nSee the [Imageomics policy for licensing](https://docs.google.com/document/d/1SlITG-r7kdJB6C8f4FCJ9Z7o7ccwldZoSRJKjhRAWVA/edit#heading=h.c1sxg0wsiqru) for more information.\n\nSee more options for the above information by clicking \"edit dataset card\" on your repo.\n\nFill in as much information as you can at each location that says \"More information needed\".\n-->\n\n<!--\nImage with caption (jpg or png):\n|![Figure #](https://huggingface.co/datasets/imageomics/<data-repo>/resolve/main/<filepath>)|\n|:--|\n|**Figure #.** [Image of <>](https://huggingface.co/datasets/imageomics/<data-repo>/raw/main/<filepath>) <caption description>.|\n-->\n\n<!--\nNotes on styling:\n\nTo render LaTex in your README, wrap the code in `\\\\(` and `\\\\)`. Example: \\\\(\\frac{1}{2}\\\\)\n\nEscape underscores (\"_\") with a \"\\\". Example: image\\_RGB\n-->\n\n# Dataset Card for Upper Waiākea Forest Reserve Passive Recordings\n\n<!-- Provide a quick summary of what the dataset is or can be used for. --> \n\n## Dataset Details\n\n### Dataset Description\n- **Curated by:** Namrata Banerji, Ekaterina Nepovinnykh, Jacob Beattie, Hikaru Keebler, Elizabeth Campolongo, Leonardo Teixeira Viotti, Amanda Navine, Patrick Hart, Tanya Berger-Wolf, and Kaiya Provost\n<!-- Provide the basic links for the dataset. These will show up on the sidebar to the right of your dataset card (\"Curated by\" too). -->\n- **Repository:** [https://github.com/Imageomics/amakiki-project]\n\n## Dataset Description\nThis is a dataset containing unlabelled, unprocessed passive acoustic recordings of Hawaiian birds in the [Upper Waiākea Forst Reserve in Hawaii](https://dlnr.hawaii.gov/forestry/frs/reserves/hawaii-island/upper-waiakea/). This dataset is intended for use in unsupervised audio analysis methods, classification using existing models, and other machine learning and ecology research purposes. \n\n### Supported Tasks and Leaderboards\nThis dataset contains passive acoustic recordings collected as part of the Fall 2024/Spring 2025 Experiential Introduction to AI and Ecology course through the [Imageomics Institute](https://github.com/imageomics) and [ABC Global Center](https://www.abcresearchcenter.org/) during January 2025.\n\nThis dataset is intended for use with unsupervised computer vision or acoustic machine learning models. No labels are provided, but recorder locations and recording timestamps are included, allowing for analysis of the relationship between ecological factors and variations in birdsong.\n\nThe dataset contains ~1623 hours of recording from 19 different recorders located in the Upper Waiākea F","encoding":"utf8","key":"README.md","name":"README.md","size":16594,"truncated":true}],"sourceObjectKey":"kipuka_metadata.csv","starterCell":"import pandas as pd\ndf = pd.read_csv(\"hf://datasets/imageomics/upper-waiakea-PAM/kipuka_metadata.csv\")\ndf.head()","starterSamplePath":"2MM00471/2MM00471_Summary.txt","totalObjects":1698},"summary":"About 1,623 hours of unlabelled passive acoustic monitoring (PAM) audio from 19 SongMeter Micro 2 recorders in the Upper Waiakea Forest Reserve on Hawaii Island, collected in…","title":"Upper Waiakea Forest Reserve Passive Acoustic Recordings of Hawaiian Forest Birds","updatedAt":"2026-11-16T17:12:31.677Z","visibility":"public","websiteUrl":"https://huggingface.co/datasets/imageomics/upper-waiakea-PAM"}}