Your challenge in this competition is to identify which species (birds, amphibians, mammals, insects) are calling in recordings made in [El Silencio Natural Reserve](https://www.fundacionbiodiversa.org/fundacion2024/information/), Colombia. This is an important task for scientists who monitor animal populations for conservation purposes. More accurate solutions could enable more comprehensive monitoring.

This competition uses a hidden test set. When your submitted notebook is scored, the actual test data will be made available to your notebook.


##Files

**train_audio/** The training data consists of short recordings of individual bird, amphibian, mammal and insects sounds generously uploaded by users of [xeno-canto.org](https://www.xeno-canto.org/), [iNaturalist](https://www.inaturalist.org) and the Colombian Sound Archive (CSA) of the [Humboldt Institute for Biological Resources Research in Colombia](https://www.humboldt.org.co). These files have been resampled to 32 kHz where applicable to match the test set audio and converted to the `ogg` format. Filenames consist of `[collection][file_id_in_collection].ogg`. The training data should have nearly all relevant files; we expect there is no benefit to looking for more on [xeno-canto.org](https://www.xeno-canto.org/) or [iNaturalist](https://www.inaturalist.org) and appreciate your cooperation in limiting the burden on their servers. If you do, please make sure to adhere to the scraping rules of these data portals.

**test_soundscapes/** When you submit a notebook, the **test_soundscapes** directory will be populated with approximately 700 recordings to be used for scoring. They are 1 minute long and in `ogg` audio format, resampled to 32 kHz. The file names are randomized, but have the general form of `soundscape_xxxxxx.ogg`. It should take your submission notebook approximately five minutes to load all the test soundscapes. Not all species from the train data actually occur in the test data.

**train_soundscapes/** Unlabeled audio data from the same recording locations as the test soundscapes. Filenames consist of `[site]_[date]_[local_time].ogg`; although recorded at the same location, precise recording sites of unlabeled soundscapes do NOT overlap with recording sites of the hidden test data.

**train.csv**  A wide range of metadata is provided for the training data. The most directly relevant fields are:

*   `primary_label`: A code for the species (eBird code for birds, iNaturalist taxon ID for non-birds). You can review detailed information about the species by appending codes to eBird and iNaturalis taxon URL, such as `https://ebird.org/species/gretin1` for the Great Tinamou or `https://www.inaturalist.org/taxa/24322` for the Red Snouted Tree Frog. Not all species have their own pages; some links might fail.
* `secondary_labels`: List of species labels that have been marked by recordists to also occur in the recording. Can be incomplete.
* `latitude` & `longitude`: Coordinates for where the recording was taken. Some bird species may have local call 'dialects,' so you may want to seek geographic diversity in your training data.
*  `author`: The user who provided the recording. `Unknown` if no name was provided.
* `filename`: The name of the associated audio file.
* `rating`: Values in 1..5 (1 - low quality, 5 - high quality) provided by users of Xeno-canto; 0 implies no rating is available; iNaturalist and the CSA do not provide quality ratings.
* `collection`: Either `XC`, `iNat` or `CSA`, indicating which collection the recording was taken from. Filenames also reference the collection and the ID within that collection.

**sample_submission.csv** A valid sample submission. 
* `row_id`: A slug of `soundscape_[soundscape_id]_[end_time]` for the prediction; e.g., Segment 00:15-00:20 of 1-minute test soundscape `soundscape_12345.ogg` has row ID `soundscape_12345_20`.
* `[species_id]`: There are 206 species ID columns. You will need to predict the probability of the presence of each species for each row.

**taxonomy.csv** - Data on the different species, including iNaturalist taxon ID and class name (Aves, Amphibia, Mammalia, Insecta).

**recording_location.txt** - Some high-level information on the recording location (El Silencio Natural Reserve).