What All Do Audio Transformer Models Hear? Probing Acoustic Representations for Language Delivery and Its Structure
Open accessThe DOI matches its Crossref record; 11 of the 11 references it cites were found in a registry or an index.
Abstract
In recent times, BERT based transformer models have become an inseparable part of the 'tech stack' of text processing models. Similar progress is being observed in the speech domain with a multitude of models observing state-of-the-art results by using audio transformer models to encode speech. This begs the question of what are these audio transformer models learning. Moreover, although the standard methodology is to choose the last layer embedding for any downstream task, but is it the optimal choice? We try to answer these questions for the two recent audio transformer models, Mockingjay and wave2vec2.0. We compare them on a comprehensive set of language delivery and structure features including audio, fluency and pronunciation features. Additionally, we probe the audio models' understanding of textual surface, syntax, and semantic features and compare them to BERT. We do this over exhaustive settings for native, non-native, synthetic, read and spontaneous speech datasets
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325 · cited 2×
“Since spoken language can be considered as a combination of spoken words and language delivery, we compare representations from both models with BERT embeddings (Devlin et al., 2019) as well We compare them”
MethodResultDeep Contextualized Word Representations
Peters, M., Neumann, M., Iyyer, M., Gardner, M., Clark, C., Lee, K., & Zettlemoyer, L. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) · doi:10.18653/v1/n18-1202 · Cited by 11,983
“Librispeech: an ASR corpus based on public domain audio books,”
Panayotov, V., Chen, G., Povey, D., & Khudanpur, S. (2015) · 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2015.7178964 · Cited by 6,072
Theoretical Bases of Communicative Approaches to Second Language Teaching and Testing
CANALE, M. (1980) · Applied Linguistics · doi:10.1093/applin/1.1.1 · Cited by 4,021
(2019) · doi:10.18653/v1/p19-1356 · Cited by 1,291 · cited 1×
“number of direct objects are hence in our set of features to evaluate the spoken content (Conneau et al., 2018; Jawahar et al., 2019).”
BackgroundEfficient transformers: A survey
Tay, Y., Dehghani, M., Bahri, D., & Metzler, D. (2023) · ACM Computing Surveys · doi:10.1145/3530811 · Cited by 1,267
Efficient Transformers: A Survey
Tay, Y., Dehghani, M., Bahri, D., & Metzler, D. (2020) · arXiv · doi:10.48550/arxiv.2009.06732 · Cited by 1,267
Common Voice: A Massively-Multilingual Speech Corpus
Ardila, R., Branson, M., Davis, K., Henretty, M., Kohler, M., Meyer, J., Morais, R., Saunders, L., Tyers, F. M., & Weber, G. (2019) · arXiv · doi:10.48550/arxiv.1912.06670 · Cited by 905 · cited 1×
“For native spontaneous English speech, we use the Mozilla Common Voice dataset [2].”
Method(2019) · doi:10.18653/v1/n19-1419 · Cited by 601
What do Neural Machine Translation Models Learn about Morphology?
Belinkov, Y., Durrani, N., Dalvi, F., Sajjad, H., & Glass, J. (2017) · Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p17-1080 · Cited by 307
Cited by
Layer-Wise Analysis of a Self-Supervised Speech Representation Model
Pasad, A., Chou, J.-C., & Livescu, K. (2021) · 2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU) · doi:10.1109/asru51503.2021.9688093 · Cited by 262
Investigating Self-Supervised Front Ends for Speech Spoofing Countermeasures
Wang, X., & Yamagishi, J. (2022) · The Speaker and Language Recognition Workshop (Odyssey 2022) · doi:10.21437/odyssey.2022-14 · Cited by 138
Evidence of Vocal Tract Articulation in Self-Supervised Learning of Speech
Cho, C. J., Wu, P., Mohamed, A., & Anumanchipalli, G. K. (2023) · ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp49357.2023.10094711 · Cited by 21
“Articula-tions are also included in the scope of [8, 7], but those are hand labeled pronunciations which are discrete and lack context dependent information such as coarticulation.”
BackgroundMethodInfluential
Metadata
- Authors
- Kumar, Y., Shah, J., Shah, R. R., & Chen, C.
- Venue
- Preprints.org
- Publisher
- MDPI AG
- Published
- 2021-01-05
- OpenAlex
- W3121914243
- Cited by
- 46
- References
- 11
- Type
- posted content
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: What All Do Audio Transformer Models Hear? Probing Acoustic Representations for Language Delivery and Its Structure. https://app.cochat.ai/papers/10.20944/preprints202101.0081.v1
- The paper
- Kumar, Y., Shah, J., Shah, R. R., & Chen, C. (2021). What All Do Audio Transformer Models Hear? Probing Acoustic Representations for Language Delivery and Its Structure. Preprints.org. https://doi.org/10.20944/preprints202101.0081.v1
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.