---
title: "WordVoice：面向LLM语音合成的显式解耦多维词级控制框架"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.06461v1"
permalink: "https://feed.astralor.com/items/itm_c26babd2aa9659b188e56ae3"
published: "2026-07-07T16:22:59.000Z"
score: 75
channel: "tech"
tags: ["LLM","TTS","WordVoice","WordVoice-5A","词级控制","韵律规划"]
---

# WordVoice：面向LLM语音合成的显式解耦多维词级控制框架

## AI 点评

> 这篇文章值得关注，因为它直接解决了LLM-TTS走向实用化的关键痛点——细粒度控制。对于从事语音合成、有声书制作、视频配音的开发者或研究者而言，WordVoice提供的词级解耦控制能力意味着可以在不牺牲自然度的前提下精确调节语速、重音、情感等属性。编辑认为，该方法在工业界落地潜力很大，尤其是结合LLM本身的语言理解能力，未来有望实现更自然的韵律自适应。

## 速览

近年基于大语言模型（LLM）的语音合成（TTS）虽自然度显著提升，但依赖隐式端到端生成，控制粒度粗糙。在音频书旁白、视频配音等需精确风格干预与严格时间对齐的场景中，无法显式操控词级声学属性成为关键瓶颈。为此，本文提出WordVoice统一框架：首先构建WordVoice-5A数据集——通过语言学引导流程标注4.7K小时双语数据，涵盖时长、边界、能量、音高和语调五维词级标注；其次设计bound-token机制将隐式生成转为显式可控范式，实现自适应多任务韵律规划与灵活手动干预；最后增强token到波形阶段的细粒度声学调制模块，弥合高压缩离散token与连续波形间的分辨率差距。实验表明，WordVoice在零样本合成稳定性保持竞争力同时，实现了多声学维度的优越解耦控制。代码与音频样本已公开。

## 关键点

- 构建 WordVoice-5A 数据集：4.7K小时双语，五维词级声学标注。
- 提出 bound-token 机制实现显式声学规划，支持多任务韵律控制与人工干预。
- 引入细粒度声学调制模块，对齐离散 token 与连续波形的词级属性。
- 实验证明多维度解耦控制性能优越，零样本合成稳定性保持竞争力。

## 全文

While recent Large Language Model (LLM)-based Text-to-Speech (TTS) systems have achieved remarkable naturalness, they predominantly rely on implicit end-to-end generation paradigms, resulting in coarse-grained control. In scenarios demanding precise stylistic interventions and strict temporal alignment, such as audiobook narration and video dubbing, the inability to explicitly manipulate word-level acoustic attributes remains a critical bottleneck. This limitation is primarily amplified by the severe scarcity of fine-grained annotated datasets and the architectural challenge of integrating multi-dimensional control signals into discrete autoregressive generation. To address this, we propose a unified framework for highly precise word-level control. First, we construct WordVoice-5A, a massive 4.7k-hour bilingual dataset featuring five-dimensional word-level annotations (duration, boundary, energy, pitch and tone) developed through a rigorous linguistically-guided pipeline. Second, we introduce WordVoice to transform the implicit generation process into an explicit, highly controllable paradigm. Specifically, we introduce a bound-token mechanism within the LLM to formulate an explicit ``acoustic planning'' process, enabling adaptive multi-task prosodic planning and flexible manual intervention. Furthermore, we augment the token-to-waveform stage with a fine-grained acoustic modulation module, bridging the resolution gap to strictly align word-level attributes between highly compressed discrete tokens and continuous waveforms. Extensive experiments demonstrate that WordVoice achieves superior, decoupled control over multiple acoustic dimensions while maintaining competitive zero-shot synthesis stability. The code and audio samples are publicly available at https://xxh333.github.io/wordvoice-demo/.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_c26babd2aa9659b188e56ae3