English

Self-Supervised Speech Models Encode Phonetic Context via Position-dependent Orthogonal Subspaces

Audio and Speech Processing 2026-03-16 v1 Computation and Language Machine Learning Sound

Abstract

Transformer-based self-supervised speech models (S3Ms) are often described as contextualized, yet what this entails remains unclear. Here, we focus on how a single frame-level S3M representation can encode phones and their surrounding context. Prior work has shown that S3Ms represent phones compositionally; for example, phonological vectors such as voicing, bilabiality, and nasality vectors are superposed in the S3M representation of [m]. We extend this view by proposing that phonological information from a sequence of neighboring phones is also compositionally encoded in a single frame, such that vectors corresponding to previous, current, and next phones are superposed within a single frame-level representation. We show that this structure has several properties, including orthogonality between relative positions, and emergence of implicit phonetic boundaries. Together, our findings advance our understanding of context-dependent S3M representations.

Keywords

Cite

@article{arxiv.2603.12642,
  title  = {Self-Supervised Speech Models Encode Phonetic Context via Position-dependent Orthogonal Subspaces},
  author = {Kwanghee Choi and Eunjung Yeo and Cheol Jun Cho and David R. Mortensen and David Harwath},
  journal= {arXiv preprint arXiv:2603.12642},
  year   = {2026}
}

Comments

Submitted to Interspeech 2026

R2 v1 2026-07-01T11:17:53.076Z