Manual annotation of individual ovenbirds in acoustic recordings
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
This dataset includes acoustic recordings and annotations of those recordings for individually identified Ovenbird songs (Seiurus aurocapilla) created from annotating randomly selected passive acoustic monitoring recordings, largely collected in Alberta, CA. Every Ovenbird song within each recording is tagged in time and frequency. Each Ovenbird tag is labeled with individual identity a number corresponding to each individual’s order of appearance in the recording. If no Ovenbirds were not detected in a given recording, the recording was labeled at approximately 2 seconds into the recording with a three-second long tag containing species label “NONE.” Recordings contain other species’ songs, but no other species are tagged. This dataset was annotated via the WildTrax interface (https://wildtrax.ca/) and can also be found at https://portal.wildtrax.ca/home/aru-tasks.html?sensorId=ARU&projectId=623. These recordings were intended to be used for studies of individual ID, but can be used for a variety of other applications, including count data, song rate and vocal activity, and migratory phenology. Because each individual Ovenbird is numbered, these data can be also be transformed into count data or used to estimate individual Ovenbirds’ song rates. Data were randomly selected with respect to recording time, recording date, and recording location. Because recording time was randomly selected, the dataset can be used to estimate Ovenbird vocal activity over the course of a day. Because recordings were selected from a date range that includes dates before Ovenbirds are likely to be present in Alberta, the dataset can be used to estimate phenology of Ovenbird migration. This dataset is intended as an example open access labeled dataset to encourage development in the field of acoustic individual identification (Knight et al. In Revision for Trends in Ecology and Evolution).
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.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.026 |
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 it