Metraipolis › archive › dc_86e433bd1bdd
Iterative Self-Learning for Expressive Text-to-Speech Synthesis — arXiv:2608.15910
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.15910
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T19:01:22.223Z
- SHA-256 of the archived text
48ec444f596ccc42e56ac0126f02064ad18d9895b2ed0ff02cce3f98e24dc362- Stable URL of this page
https://metraipolis.com/documento/dc_86e433bd1bdd
How to cite
Iterative Self-Learning for Expressive Text-to-Speech Synthesis — arXiv:2608.15910 — arXiv (en), https://arxiv.org/abs/2608.15910 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_86e433bd1bdd (Metraipolis archive dc_86e433bd1bdd, sha256 48ec444f596ccc42e56ac0126f02064ad18d9895b2ed0ff02cce3f98e24dc362)
BibTeX
@misc{dc86e433bd1bdd,
title = {Iterative Self-Learning for Expressive Text-to-Speech Synthesis — arXiv:2608.15910},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.15910},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_86e433bd1bdd; sha256 48ec444f596ccc42e56ac0126f02064ad18d9895b2ed0ff02cce3f98e24dc362},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.15910. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Expressive text-to-speech (TTS) systems that use explicit conditioning labels provide direct and interpretable control over expressive attributes, in contrast to reference-based or prompting-based approaches, but require labeled data. Obtaining these labels at scale is costly and time-consuming, yet no prior semi-supervised framework addresses this specific bottleneck. Existing semi-supervised TTS methods instead target scarcity of paired speech-text data or transcriptions. To address the scarcity of expressive labels, we propose an Iterative Self-Learning (ISL) framework for expressive TTS, built on Invert-Classify, a classifier-free method that recovers discrete expressive labels by inverting a frozen generative model. The framework iteratively pseudo-labels unlabeled speech using the current model, retrains on the combined labeled and pseudo-labeled data, and repeats, progressively refining label quality and synthesis. We validate on two expressive tasks, word-level prominence and utterance-level emotion, across multiple low-resource data splits. We find that iterative refinement can improve pseudo-label accuracy over single-pass baselines. Furthermore, we observe that these improvements in pseudo-labeling of expressivity translate to gains in expressive label adherence and synthesis quality, confirmed by objective metrics and human listening tests. In the most data-scarce conditions, ISL-trained models outperform single-pass pseudo-labeling and further approach fully supervised performance, demonstrating that gradient-based ISL is an effective solution to expressive label scarcity in low-resource TTS. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.CL). Categorie arXiv: Audio and Speech Processing (eess.AS) ; Computation and Language (cs.CL); Sound (cs.SD). Pagina del lavoro: https://arxiv.org/abs/2608.15910 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-18T18:06:57.924Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).