# Exploring Timbre Disentanglement in Non-Autoregressive Cross-Lingual Text-to-Speech

arXiv: https://arxiv.org/abs/2110.07192 · HTML: https://mlsanity.com/paper/2110.07192/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Oct 2021.

## Built on by (most cited first, 1 of 1)

- [DiCLET-TTS: Diffusion Model based Cross-lingual Emotion Transfer for Text-to-Speech -- A Study between English and Mandarin](https://mlsanity.com/paper/2309.00883/) (2023)

## Abstract (arXiv)

In this paper, we study the disentanglement of speaker and language representations in non-autoregressive cross-lingual TTS models from various aspects. We propose a phoneme length regulator that solves the length mismatch problem between IPA input sequence and monolingual alignment results. Using the phoneme length regulator, we present a FastPitch-based cross-lingual model with IPA symbols as input representations. Our experiments show that language-independent input representations (e.g. IPA symbols), an increasing number of training speakers, and explicit modeling of speech variance information all encourage non-autoregressive cross-lingual TTS model to disentangle speaker and language representations. The subjective evaluation shows that our proposed model can achieve decent naturalness and speaker similarity in cross-language voice cloning.