Deep Learning for Speech Accent Detection in Videogames
Bibliographic record
Abstract
In video games, a wide range of characters make up the world players inhabit. These characters, NPCs, have traits, such as their appearance and speech accent, that determine certain things about them, including moral inclination, levels of trustworthiness, social class, levels of education, and ethnic background. But what does an accent say about a character in a video game? We use deep learning to train a neural network to detect speech accents and establish the degree to which machines can be used to recognize these accents. This research aims to help sociolinguists and discourse analysts establish critical study and content analytical findings for instance about stereotypical uses of speech accents, to better analyze who has what accent in video games, and what kind of language ideologies and social value judgments the use of accents in games construct and perpetuate. This paper presents the results of the first deep learning experiments, which were conducted on Standard North American, British Received Pronunciation, and Spanish English. We discuss our methodological considerations and some early deep learning results, which show relatively low levels of accuracy (61%). We discuss possibilities of improving our method, and of enriching our training datasets.
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 | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".