English

Multi-Task Learning for Front-End Text Processing in TTS

Computation and Language 2024-04-04 v1

Abstract

We propose a multi-task learning (MTL) model for jointly performing three tasks that are commonly solved in a text-to-speech (TTS) front-end: text normalization (TN), part-of-speech (POS) tagging, and homograph disambiguation (HD). Our framework utilizes a tree-like structure with a trunk that learns shared representations, followed by separate task-specific heads. We further incorporate a pre-trained language model to utilize its built-in lexical and contextual knowledge, and study how to best use its embeddings so as to most effectively benefit our multi-task model. Through task-wise ablations, we show that our full model trained on all three tasks achieves the strongest overall performance compared to models trained on individual or sub-combinations of tasks, confirming the advantages of our MTL framework. Finally, we introduce a new HD dataset containing a balanced number of sentences in diverse contexts for a variety of homographs and their pronunciations. We demonstrate that incorporating this dataset into training significantly improves HD performance over only using a commonly used, but imbalanced, pre-existing dataset.

Keywords

Cite

@article{arxiv.2401.06321,
  title  = {Multi-Task Learning for Front-End Text Processing in TTS},
  author = {Wonjune Kang and Yun Wang and Shun Zhang and Arthur Hinsvark and Qing He},
  journal= {arXiv preprint arXiv:2401.06321},
  year   = {2024}
}

Comments

ICASSP 2024

R2 v1 2026-06-28T14:14:51.824Z