MELA-TTS: Joint transformer-diffusion model with representation alignment for speech synthesis

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: An, Keyu, Zhang, Zhiyu, Gao, Changfeng, Li, Yabin, Peng, Zhendong, Wang, Haoxu, Du, Zhihao, Zhao, Han, Gao, Zhifu, Li, Xiangang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918304277331968
author An, Keyu
Zhang, Zhiyu
Gao, Changfeng
Li, Yabin
Peng, Zhendong
Wang, Haoxu
Du, Zhihao
Zhao, Han
Gao, Zhifu
Li, Xiangang
author_facet An, Keyu
Zhang, Zhiyu
Gao, Changfeng
Li, Yabin
Peng, Zhendong
Wang, Haoxu
Du, Zhihao
Zhao, Han
Gao, Zhifu
Li, Xiangang
contents This work introduces MELA-TTS, a novel joint transformer-diffusion framework for end-to-end text-to-speech synthesis. By autoregressively generating continuous mel-spectrogram frames from linguistic and speaker conditions, our architecture eliminates the need for speech tokenization and multi-stage processing pipelines. To address the inherent difficulties of modeling continuous features, we propose a representation alignment module that aligns output representations of the transformer decoder with semantic embeddings from a pretrained ASR encoder during training. This mechanism not only speeds up training convergence, but also enhances cross-modal coherence between the textual and acoustic domains. Comprehensive experiments demonstrate that MELA-TTS achieves state-of-the-art performance across multiple evaluation metrics while maintaining robust zero-shot voice cloning capabilities, in both offline and streaming synthesis modes. Our results establish a new benchmark for continuous feature generation approaches in TTS, offering a compelling alternative to discrete-token-based paradigms.
format Preprint
id arxiv_https___arxiv_org_abs_2509_14784
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MELA-TTS: Joint transformer-diffusion model with representation alignment for speech synthesis
An, Keyu
Zhang, Zhiyu
Gao, Changfeng
Li, Yabin
Peng, Zhendong
Wang, Haoxu
Du, Zhihao
Zhao, Han
Gao, Zhifu
Li, Xiangang
Audio and Speech Processing
This work introduces MELA-TTS, a novel joint transformer-diffusion framework for end-to-end text-to-speech synthesis. By autoregressively generating continuous mel-spectrogram frames from linguistic and speaker conditions, our architecture eliminates the need for speech tokenization and multi-stage processing pipelines. To address the inherent difficulties of modeling continuous features, we propose a representation alignment module that aligns output representations of the transformer decoder with semantic embeddings from a pretrained ASR encoder during training. This mechanism not only speeds up training convergence, but also enhances cross-modal coherence between the textual and acoustic domains. Comprehensive experiments demonstrate that MELA-TTS achieves state-of-the-art performance across multiple evaluation metrics while maintaining robust zero-shot voice cloning capabilities, in both offline and streaming synthesis modes. Our results establish a new benchmark for continuous feature generation approaches in TTS, offering a compelling alternative to discrete-token-based paradigms.
title MELA-TTS: Joint transformer-diffusion model with representation alignment for speech synthesis
topic Audio and Speech Processing
url https://arxiv.org/abs/2509.14784