Discrete Unit-based Low-latency Multi-lingual Speech Synthesis for LIMMITS’25 Challenge

In this paper, we present the system developed by our team, CCATTS, for the LIMMITS’25 challenge, focusing on few-shot and zero-shot TTS. We adopt a two-stage TTS strategy. In track 1, we fine-tune the pre-trained ZMM-TTS model and successfully achieve multilingual low-latency TTS. In track 2, we propose a token-based framework by modifying the first-stage model of ZMM-TTS, disentangling speech features into four types of discrete tokens—Content, Acoustic, Emotion, and Speaker—and integrating it with our designed token2wav module. This module consists of a HiFi-GAN-style decoder, an acoustic refiner, and U-Net flow matching, to generate high-quality speech. The official competition results demonstrate that our method achieves strong performance in both tracks.

Authors

Institutions

Publication Details

Published
2025-03-12
DOI
https://doi.org/10.1109/icassp49660.2025.10890527
Primary Topic
Speech Recognition and Synthesis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Discrete Unit-based Low-latency Multi-lingual Speech Synthesis for LIMMITS’25 Challenge

Yuheng Lu, Jianwu Dang, Chunyu Qiang, Haoyu Wang et al.
Speech Recognition and Synthesis
article

Discrete Unit-based Low-latency Multi-lingual Speech Synthesis for LIMMITS’25 Challenge

Yuheng Lu, Jianwu Dang, Chunyu Qiang, Haoyu Wang, Tianrui Wang, Qiuyu Liu, Gong Cheng, Yu Jiang, Xiao-Lei Zhang, Longbiao Wang, Xiaobao Wang
article en

Abstract

In this paper, we present the system developed by our team, CCATTS, for the LIMMITS’25 challenge, focusing on few-shot and zero-shot TTS. We adopt a two-stage TTS strategy. In track 1, we fine-tune the pre-trained ZMM-TTS model and successfully achieve multilingual low-latency TTS. In track 2, we propose a token-based framework by modifying the first-stage model of ZMM-TTS, disentangling speech features into four types of discrete tokens—Content, Acoustic, Emotion, and Speaker—and integrating it with our designed token2wav module. This module consists of a HiFi-GAN-style decoder, an acoustic refiner, and U-Net flow matching, to generate high-quality speech. The official competition results demonstrate that our method achieves strong performance in both tracks.

Tianjin University (CN), Chinese Academy of Sciences (CN), China Telecom (China) (CN), Shenzhen Institutes of Advanced Technology (CN), China Telecom (CN)
Openalex Percentile: Top 98%
Speech Recognition and Synthesis
DOISourceOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Discrete Unit-based Low-latency Multi-lingual Speech Synthesis for LIMMITS’25 Challenge — Yuheng Lu, Jianwu Dang, et al. · (2025) | TGRS Research Map | TGRS