Multimodal and Spectral Degradation Effects on Speech and Emotion Recognition in Adults
Bibliographic record
Abstract
Research has shown that individuals with severe hearing loss demonstrate considerable adaptation to hearing input from their cochlear implants (CIs), especially when implanted at younger ages. Despite these gains, hearing restoration with sensory prostheses does not match that of normal acoustic hearing. Limitations are especially apparent in complex listening situations. CIs retain important timing information, but discard fine pitch details that are informative to voice quality and music. We examined how speech and emotion recognition can be improved for CI listeners by the addition of informative multimodal (auditory and visual) cues. We created conditions that simulate the hearing experiences of CI listeners using a vocoder, which reduced the fine pitch information. In the unimodal auditory condition, hearing adult participants listened to sentence-length vocoded speech created with 4, 8, 16, and 32 bands that contained increasing amounts of spectral (pitch) detail, respectively. In the multimodal condition, the vocoded speech was superimposed to videos of the talker speaking the sentence. Our results show that listeners capitalized on informative visual cues that complemented the acoustic information and improved their speech and emotion recognition accuracies. The multimodal benefit was greater under the most difficult listening conditions; that is, 4 and 8 bands. Our findings also show that the addition of visual information benefited emotion recognition more greatly, where spectral degradation hampered the perception of important prosodic detail that cue emotion in voice. The present findings can be used to inform rehabilitative practices by incorporating informative multimodal cues to improve communication outcomes of CI listeners. Discipline: Psychology Honours Faculty Mentor: Dr. Tara Vongpaisal
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 | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.003 | 0.000 |
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".