Quantifying Dimensional Independence in Speech: An Information-Theoretic Framework for Disentangled Representation Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910031335653376 |
|---|---|
| author | Kashyap, Bipasha Schuller, Björn W. Pathirana, Pubudu N. |
| author_facet | Kashyap, Bipasha Schuller, Björn W. Pathirana, Pubudu N. |
| contents | Speech signals encode emotional, linguistic, and pathological information within a shared acoustic channel; however, disentanglement is typically assessed indirectly through downstream task performance. We introduce an information-theoretic framework to quantify cross-dimension statistical dependence in handcrafted acoustic features by integrating bounded neural mutual information (MI) estimation with non-parametric validation. Across six corpora, cross-dimension MI remains low, with tight estimation bounds ($< 0.15$ nats), indicating weak statistical coupling in the data considered, whereas Source--Filter MI is substantially higher (0.47 nats). Attribution analysis, defined as the proportion of total MI attributable to source versus filter components, reveals source dominance for emotional dimensions (80\%) and filter dominance for linguistic and pathological dimensions (60\% and 58\%, respectively). These findings provide a principled framework for quantifying dimensional independence in speech. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_20592 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Quantifying Dimensional Independence in Speech: An Information-Theoretic Framework for Disentangled Representation Learning Kashyap, Bipasha Schuller, Björn W. Pathirana, Pubudu N. Sound Audio and Speech Processing Speech signals encode emotional, linguistic, and pathological information within a shared acoustic channel; however, disentanglement is typically assessed indirectly through downstream task performance. We introduce an information-theoretic framework to quantify cross-dimension statistical dependence in handcrafted acoustic features by integrating bounded neural mutual information (MI) estimation with non-parametric validation. Across six corpora, cross-dimension MI remains low, with tight estimation bounds ($< 0.15$ nats), indicating weak statistical coupling in the data considered, whereas Source--Filter MI is substantially higher (0.47 nats). Attribution analysis, defined as the proportion of total MI attributable to source versus filter components, reveals source dominance for emotional dimensions (80\%) and filter dominance for linguistic and pathological dimensions (60\% and 58\%, respectively). These findings provide a principled framework for quantifying dimensional independence in speech. |
| title | Quantifying Dimensional Independence in Speech: An Information-Theoretic Framework for Disentangled Representation Learning |
| topic | Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2602.20592 |