Research

Paper

AI LLM March 17, 2026

CAST-TTS: A Simple Cross-Attention Framework for Unified Timbre Control in TTS

Authors

Zihao Zheng, Wen Wu, Chao Zhang, Mengyue Wu, Xuenan Xu

Abstract

Current Text-to-Speech (TTS) systems typically use separate models for speech-prompted and text-prompted timbre control. While unifying both control signals into a single model is desirable, the challenge of cross-modal alignment often results in overly complex architectures and training objective. To address this challenge, we propose CAST-TTS, a simple yet effective framework for unified timbre control. Features are extracted from speech prompts and text prompts using pre-trained encoders. The multi-stage training strategy efficiently aligns the speech and projected text representations within a shared embedding space. A single cross-attention mechanism then allows the model to use either of these representations to control the timbre. Extensive experiments validate that the unified cross-attention mechanism is critical for achieving high-quality synthesis. CAST-TTS achieves performance comparable to specialized single-input models while operating within a unified architecture. The demo page can be accessed at https://HiRookie9.github.io/CAST-TTS-Page.

Metadata

arXiv ID: 2603.16280
Provider: ARXIV
Primary Category: cs.SD
Published: 2026-03-17
Fetched: 2026-03-18 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.16280v1</id>\n    <title>CAST-TTS: A Simple Cross-Attention Framework for Unified Timbre Control in TTS</title>\n    <updated>2026-03-17T09:11:24Z</updated>\n    <link href='https://arxiv.org/abs/2603.16280v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.16280v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Current Text-to-Speech (TTS) systems typically use separate models for speech-prompted and text-prompted timbre control. While unifying both control signals into a single model is desirable, the challenge of cross-modal alignment often results in overly complex architectures and training objective. To address this challenge, we propose CAST-TTS, a simple yet effective framework for unified timbre control. Features are extracted from speech prompts and text prompts using pre-trained encoders. The multi-stage training strategy efficiently aligns the speech and projected text representations within a shared embedding space. A single cross-attention mechanism then allows the model to use either of these representations to control the timbre. Extensive experiments validate that the unified cross-attention mechanism is critical for achieving high-quality synthesis. CAST-TTS achieves performance comparable to specialized single-input models while operating within a unified architecture. The demo page can be accessed at https://HiRookie9.github.io/CAST-TTS-Page.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.SD'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='eess.AS'/>\n    <published>2026-03-17T09:11:24Z</published>\n    <arxiv:comment>Submitted to Interspeech 2026</arxiv:comment>\n    <arxiv:primary_category term='cs.SD'/>\n    <author>\n      <name>Zihao Zheng</name>\n    </author>\n    <author>\n      <name>Wen Wu</name>\n    </author>\n    <author>\n      <name>Chao Zhang</name>\n    </author>\n    <author>\n      <name>Mengyue Wu</name>\n    </author>\n    <author>\n      <name>Xuenan Xu</name>\n    </author>\n  </entry>"
}