UR-BERT: Scaling Text Encoders for Massively Multilingual TTS Through Universal Romanization and Speech Token Prediction

Sangmin Lee, Eekgyun Ahn, Woongjib Choi, Hong-Goo Kang
6/10/2026
cs.CLcs.SD

No confident code implementation found for this paper. 1 possible match can be shown below.

Abstract

We propose UR-BERT, a Romanized transcription-based text-to-speech (TTS) encoder for massively multilingual TTS systems. Conventional grapheme-to-phoneme (G2P)-based approaches are limited to around 100 languages due to the availability of reliable G2P resources. In contrast, UR-BERT scales to 495 languages by unifying diverse writing systems into a shared Romanization representation. To further enhance phonetic fidelity and text-speech alignment, we introduce a speech token prediction objective during training, which encourages the encoder to learn speech-aware phonetic representations in a data-efficient manner. Experiments show that TTS systems built on UR-BERT consistently outperform recent text encoder baselines across a wide range of languages and resource conditions, and demonstrate strong generalization to unseen languages.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. 1 weaker match is hidden by default — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Omnilingual SONAR: Cross-Lingual and Cross-Modal Sentence Embeddings Bridging Massively Multilingual Text and Speech

86% match2026

Zero-AVSR: Zero-Shot Audio-Visual Speech Recognition with LLMs by Learning Language-Agnostic Speech Representations

85% match1 repo362025

SONAR: Sentence-Level Multimodal and Language-Agnostic Representations

85% match3 repos9132023

Textless Unit-to-Unit training for Many-to-Many Multilingual Speech-to-Speech Translation

85% match1 repo312023

Towards a Deep Understanding of Multilingual End-to-End Speech Translation

85% match1 repo52023

SeamlessM4T: Massively Multilingual & Multimodal Machine Translation

85% match3 repos11.9k2023

Cite this paper

@article{lee2026urbert,
  title  = {UR-BERT: Scaling Text Encoders for Massively Multilingual TTS Through Universal Romanization and Speech Token Prediction},
  author = {Sangmin Lee and Eekgyun Ahn and Woongjib Choi and Hong-Goo Kang},
  year   = {2026},
  eprint = {2606.11681},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2606.11681v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2606.11681.svg)](https://codeofpaper.com/paper/2606.11681)

Discussion