Self-Supervised Speech Models Encode Phonetic Context via Position-dependent Orthogonal Subspaces
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914390537666560 |
|---|---|
| author | Choi, Kwanghee Yeo, Eunjung Cho, Cheol Jun Mortensen, David R. Harwath, David |
| author_facet | Choi, Kwanghee Yeo, Eunjung Cho, Cheol Jun Mortensen, David R. Harwath, David |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_12642 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Self-Supervised Speech Models Encode Phonetic Context via Position-dependent Orthogonal Subspaces Choi, Kwanghee Yeo, Eunjung Cho, Cheol Jun Mortensen, David R. Harwath, David Audio and Speech Processing Computation and Language Machine Learning Sound 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. |
| title | Self-Supervised Speech Models Encode Phonetic Context via Position-dependent Orthogonal Subspaces |
| topic | Audio and Speech Processing Computation and Language Machine Learning Sound |
| url | https://arxiv.org/abs/2603.12642 |