A Technique for Isolating Lexically-Independent Phonetic Dependencies in Generative CNNs
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912424503803904 |
|---|---|
| author | Šegedin, Bruno Ferenc |
| author_facet | Šegedin, Bruno Ferenc |
| contents | The ability of deep neural networks (DNNs) to represent phonotactic generalizations derived from lexical learning remains an open question. This study (1) investigates the lexically-invariant generalization capacity of generative convolutional neural networks (CNNs) trained on raw audio waveforms of lexical items and (2) explores the consequences of shrinking the fully-connected layer (FC) bottleneck from 1024 channels to 8 before training. Ultimately, a novel technique for probing a model's lexically-independent generalizations is proposed that works only under the narrow FC bottleneck: generating audio outputs by bypassing the FC and inputting randomized feature maps into the convolutional block. These outputs are equally biased by a phonotactic restriction in training as are outputs generated with the FC. This result shows that the convolutional layers can dynamically generalize phonetic dependencies beyond lexically-constrained configurations learned by the FC. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_09218 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Technique for Isolating Lexically-Independent Phonetic Dependencies in Generative CNNs Šegedin, Bruno Ferenc Computation and Language Sound Audio and Speech Processing The ability of deep neural networks (DNNs) to represent phonotactic generalizations derived from lexical learning remains an open question. This study (1) investigates the lexically-invariant generalization capacity of generative convolutional neural networks (CNNs) trained on raw audio waveforms of lexical items and (2) explores the consequences of shrinking the fully-connected layer (FC) bottleneck from 1024 channels to 8 before training. Ultimately, a novel technique for probing a model's lexically-independent generalizations is proposed that works only under the narrow FC bottleneck: generating audio outputs by bypassing the FC and inputting randomized feature maps into the convolutional block. These outputs are equally biased by a phonotactic restriction in training as are outputs generated with the FC. This result shows that the convolutional layers can dynamically generalize phonetic dependencies beyond lexically-constrained configurations learned by the FC. |
| title | A Technique for Isolating Lexically-Independent Phonetic Dependencies in Generative CNNs |
| topic | Computation and Language Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2506.09218 |