The Reduction of Speech Characteristic Vector Using PSO Algorithm and the Evaluation of the Effectiveness of Different Speech Characteristic in Recognition of Persian Language State
Bibliographic record
Abstract
Speech has a number of characteristic, the extraction of which can play an important role in the accuracy of speech recognition. In this regard, many researchers have attempted to investigate these features and provide methods which enhance the recognition and identification of speech states. These features include MFCC coefficients, energy, Formant Frequency and Pitch Frequency which are highly important in speech state recognition system. This paper explores the effect of these features on speech state recognition and four different states, i.e. angry, happy, natural and question will be tested. The study investigates a variety of speech characteristics in form of a vector contains 55-characteristics. In the next step, drawing on PSO optimization algorithm, 49, 24 and 15-characteristic vectors are achieved. The less the characteristics of a vector are, the higher the action velocity will be. After that, the mean Normalization, Cepstral variance and Cepstral gain methods are applied on these vectors and using GMM algorithm, speech state recognition is executed on normalized vectors. Finally, following the normalization of the output vectors and speech state recognition through GMM algorithm, the effect of different speech characteristics as well as different normalization methods on speech state recognition are examined.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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".