MétaCan
Menu
Back to cohort

Study for Automatic Speech Recognition for Wav2Vec2.0

2025· article· en· W4410617847 on OpenAlex
Xiwei Huang

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueApplied and Computational Engineering · 2025
Typearticle
Languageen
FieldComputer Science
TopicSpeech Recognition and Synthesis
Canadian institutionsUniversity of Toronto
Fundersnot available
KeywordsSpeech recognitionComputer scienceNatural language processing

Abstract

fetched live from OpenAlex

Automatic Speech Recognition (ASR) is a popular technology that converts speech audio into corresponding text. This application serves critical roles in areas such as virtual assistants, transcription services, accessibility tools, etc. This paper mainly introduces the application of the Wav2Vec2.0 model, which is an advanced self-supervised ASR model. The dataset used in this research is the Mozilla Common Voice dataset, which contains audio data in multiple languages and from people across different ages, genders, and occupations. In addition, the data preprocessing process and the architecture of the model will also be discussed in this research. The implementation demonstrates the strong ability of the Wav2Vec2.0 model in transcribing speech data from the Mozilla Common Voice database, and the experimental results highlight the model's robustness in handling variations in accent, speaking speed, and recording quality, achieving competitive word error rates (WER) across diverse linguistic scenarios. Results also indicate potential improvements in accuracy through more careful and targeted data processing and improving the tokenizer. All these findings underscore the model's future capacity in real-world speech recognition systems, emphasizing its adaptability and efficiency.

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.

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 categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Other design · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.943
Threshold uncertainty score0.399

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 communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.019
GPT teacher head0.248
Teacher spread0.228 · 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