CrossAccent-TTS: Cross-Lingual Accent-Intensity Controllable Text-to-Speech via Disentangled Speaker and Accent Representations

2026-06-24

AI Summary

This work addresses the challenge of explicitly controlling accent characteristics and their intensity in cross-lingual text-to-speech synthesis for low-resource, phonologically diverse Indic languages. The authors propose a method that disentangles speaker and accent representations by introducing an Accent Intensity Controller (AIC) and a plug-in accent subspace mechanism that weights language embeddings. Integrated within an end-to-end TTS framework, this approach enables both cross-lingual accent transfer and fine-grained intensity modulation. Notably, it allows for continuous adjustment of accent strength during inference while preserving speaker identity and speech naturalness. Experiments on the Indic Multilingual and L2-arctic datasets demonstrate that the proposed model significantly outperforms existing baselines in terms of accent similarity, controllability, speaker consistency, and overall naturalness.

Abstract

Accent conversion and controllability remain fundamental challenges in cross-lingual text-to-speech (TTS), particularly for low-resource and phonetically diverse Indic languages. While recent large language model (LLM)-based TTS systems exhibit strong cross-lingual generalization, they provide limited explicit control over accent characteristics and intensity. In this paper, we propose CrossAccentTTS, a framework that enables both accent control and conversion while preserving speaker identity. Specifically, we introduce an Accent Intensity Controller (AIC) that injects weighted language embeddings into the accent subspace, allowing smooth interpolation between accents and fine-grained modulation of accent strength at inference time. Experiments on the Indic Multilingual and L2-arctic datasets shows that CrossAccent-TTS achieves precise control of accent intensity, outperforming strong baselines in accent similarity and controllability by maintaining speaker similarity and naturalness.

Problem

• accent control
• cross-lingual TTS
• accent intensity
• speaker identity preservation
• low-resource languages

Innovation

• accent control
• cross-lingual TTS
• disentangled representation
• accent intensity modulation
• speaker identity preservation