Evaluating the Representation of Vowels in Wav2Vec Feature Extractor: A Layer-Wise Analysis Using MFCCs
Not on the map. Why?
Abstract (arXiv)
Automatic Speech Recognition has advanced with self-supervised learning, enabling feature extraction directly from raw audio. In Wav2Vec, a CNN first transforms audio into feature vectors before the transformer processes them. This study examines CNN-extracted information for monophthong vowels using the TIMIT corpus. We compare MFCCs, MFCCs with formants, and CNN activations by training SVM classifiers for front-back vowel identification, assessing their classification accuracy to evaluate phonetic representation.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 7 of its 19 references that are in the map. First posted on arXiv 25 Aug 2025.
Builds on
7 of 19 references.
References · 7
- 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,905
- 2021Layer-wise Analysis of a Self-supervised Speech Representation Modelcited by 298
- 2021Simple and Effective Zero-shot Cross-lingual Phoneme Recognitioncited by 79
- 2017Analyzing Hidden Representations in End-to-End Automatic Speech Recognition Systemscited by 32
- 2020Probing Acoustic Representations for Phonetic Propertiescited by 26
- 2019Analyzing Phonetic and Graphemic Representations in End-to-End Automatic Speech Recognitioncited by 16
- 2023Comparative Analysis of the wav2vec 2.0 Feature Extractorcited by 4
Built on by
Not cited yet.