MétaCan
Menu
Back to cohort

Transformer-Based Multi-Head Attention for Noisy Speech Recognition

2024· article· en· W4407901436 on OpenAlexaff
Noussaiba Djeffal, Hamza Kheddar, Djamel Addou, Sid‐Ahmed Selouani

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicSpeech Recognition and Synthesis
Canadian institutionsUniversité de Moncton
Fundersnot available
KeywordsComputer scienceSpeech recognitionTransformerArtificial intelligenceEngineeringElectrical engineeringVoltage

Abstract

fetched live from OpenAlex

This paper presents an advanced transformer-based multi-head attention approach for automatic speech recognition (ASR) in noisy speech environments. The study focuses on four distinct noise scenarios: subway, babble, car, and exhibition hall, each rigorously evaluated across different signal-to-noise ratios (SNRs) and a clean condition. Utilizing the AURORA 2 dataset, this research investigates the performance and robustness of the proposed model under these varied acoustic conditions. To further enhance the transformer-based attention’s performance in ASR, we compare its effectiveness against sound classification in environmental noise using the ESC-50 dataset. The experimental results provide a comprehensive analysis, demonstrating that the model with transformer-based multi-head attention, incorporating merging and gammatone frequency cepstral coefficients (GFCC) features, significantly outperforms a baseline model using GFCC features with transformer-based multi-head attention alone, without merging. The model achieves an accuracy of 98.75% on the AURORA 2 dataset in clean conditions, and maintains a high accuracy of 92.38% under noisy conditions. When evaluated on the ESC-10 dataset, it achieves an accuracy of 81.25%, underscoring its robustness and potential for practical deployment in diverse acoustic settings.

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 distilled prediction

Teacher imitation

Not 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.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesInsufficient payload (model declined to judge)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Other design · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.989
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.001
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.001

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.

Opus teacher head0.084
GPT teacher head0.316
Teacher spread0.232 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

Study designOther design
Domainnot available
GenreMethods

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".

Quick stats

Citations0
Published2024
Admission routes1
Has abstractyes

Explore more

Same topicSpeech Recognition and SynthesisFrench-language works237,207