The phonetically balanced kindergarten assessment: An alternative evaluation method utilizing machine learning technology
Bibliographic record
Abstract
This study investigates the use case of machine learning (ML) technologies to score the Phonetically Balanced Kindergarten (PBK) word test. The PBK, developed to test pediatric speech perception, is also used to monitor outcomes of cochlear implantation (CI) in children. Many factors contribute to high PBK score variability in CI users, but effects of scoring by human listeners are unclear. In this study, an alternate ML scoring method was developed. The Ursa classifier (Speechmatics, 2023) was used to recognize 100 PBK words spoken by 12 adults with normal hearing, speech, and language. The spoken words (n = 1200) were manipulated in six conditions (first and last phoneme deletion; high frequency filtering at 1, 2, 4, and 6 kHz), yielding 8400 stimuli. The classifier correctly recognized most of the unaltered words (true positive rate of 0.90; peak distribution density at 96%) and provided a null or incorrect response to altered stimuli as expected; most frequently for the first phoneme deletion and 1 kHz filtered conditions. Inter-class correlations showed comparable results between Ursa and 7 normal hearing human scorers (Cohens Kappa: 0.39). Thus, ML tools show promise in the scoring of the PBK; increasing the accuracy, efficiency, and reliability of this clinical assessment.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.016 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".