Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.00100 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911349944090624 |
|---|---|
| author | Yeh, Sung-Lin Bell, Peter Tang, Hao |
| author_facet | Yeh, Sung-Lin Bell, Peter Tang, Hao |
| contents | Despite being the best known objective for learning speech representations, the HuBERT objective has not been further developed and improved. We argue that it is the lack of an underlying principle that stalls the development, and, in this paper, we show that predictive coding under a variational view is the principle behind the HuBERT objective. Due to its generality, our formulation provides opportunities to improve parameterization and optimization, and we show two simple modifications that bring immediate improvements to the HuBERT objective. In addition, the predictive coding formulation has tight connections to various other objectives, such as APC, CPC, wav2vec, and BEST-RQ. Empirically, the improvement in pre-training brings significant improvements to four downstream tasks: phone classification, f0 tracking, speaker recognition, and automatic speech recognition, highlighting the importance of the predictive coding interpretation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_00100 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning Speech Representations with Variational Predictive Coding Yeh, Sung-Lin Bell, Peter Tang, Hao Audio and Speech Processing Computation and Language Despite being the best known objective for learning speech representations, the HuBERT objective has not been further developed and improved. We argue that it is the lack of an underlying principle that stalls the development, and, in this paper, we show that predictive coding under a variational view is the principle behind the HuBERT objective. Due to its generality, our formulation provides opportunities to improve parameterization and optimization, and we show two simple modifications that bring immediate improvements to the HuBERT objective. In addition, the predictive coding formulation has tight connections to various other objectives, such as APC, CPC, wav2vec, and BEST-RQ. Empirically, the improvement in pre-training brings significant improvements to four downstream tasks: phone classification, f0 tracking, speaker recognition, and automatic speech recognition, highlighting the importance of the predictive coding interpretation. |
| title | Learning Speech Representations with Variational Predictive Coding |
| topic | Audio and Speech Processing Computation and Language |
| url | https://arxiv.org/abs/2601.00100 |