Research Papers 论文研究 3d ago Updated 3d ago 更新于 3天前 49

GRAFT: Grafted Reference Audio for Fine-grained Pronunciation in Zero-shot Text-to-Speech GRAFT:用于零样本语音合成中细粒度发音的嫁接参考音频

GRAFT introduces a per-word pronunciation conditioning mechanism for neural codec language modeling in zero-shot text-to-speech systems. The method uses a short spoken sample of a specific word, encoded by the model’s own tokenizer, to disambiguate rare proper nouns and technical terms. Voice conversion during training disentangles the hint speaker from the target speaker, allowing pronunciation hints from any voice while maintaining the target identity. GRAFT reduces target-word phoneme error r 提出GRAFT机制,通过植入参考音频片段解决零样本TTS中罕见专有名词和技术术语的发音歧义问题。 利用模型自带的语音分词器编码参考音频,并将其绑定到提示词中的特定单词位置,实现细粒度的发音控制。 在训练数据构建阶段引入声音转换技术,解耦提示说话人与目标说话人,确保参考音频可来自任意声音而不影响最终音色一致性。 盲听测试显示人类评分者认为GRAFT生成的困难单词最接近参考录音;五语言基准测试将目标单词音素错误率降低22-39%。

65
Hot 热度
75
Quality 质量
70
Impact 影响力

Analysis 深度分析

TL;DR

  • GRAFT introduces a per-word pronunciation conditioning mechanism for neural codec language modeling in zero-shot text-to-speech systems.
  • The method uses a short spoken sample of a specific word, encoded by the model’s own tokenizer, to disambiguate rare proper nouns and technical terms.
  • Voice conversion during training disentangles the hint speaker from the target speaker, allowing pronunciation hints from any voice while maintaining the target identity.
  • GRAFT reduces target-word phoneme error rates by 22-39% compared to text-only backbones and outperforms existing open-source zero-shot systems on pronunciation accuracy.
  • Human listeners ranked GRAFT first in blind studies, judging its rendering of difficult words closest to the reference recordings.

Why It Matters

This advancement addresses a critical limitation in current zero-shot TTS systems: the inability to reliably pronounce rare or ambiguous words without extensive fine-tuning or manual phoneme annotation. By enabling precise, per-word control over pronunciation while preserving speaker identity and naturalness, GRAFT significantly enhances the utility of TTS for professional applications involving specialized terminology, names, or multilingual content.

Technical Details

  • Mechanism: GRAFT binds a spoken audio snippet of a target word to its corresponding position in the text prompt, using the model's internal speech tokenizer for encoding.
  • Training Strategy: Employs voice conversion during dataset construction to separate the pronunciation source (hint speaker) from the target voice, ensuring the model learns to transfer pronunciation characteristics independently of speaker identity.
  • Benchmarking: Evaluated on a five-language objective benchmark, demonstrating superior performance in target-word pronunciation over both phoneme-conditioned and text-conditioned baselines.
  • Performance Metrics: Achieved a 22-39% reduction in phoneme error rate for target words compared to the identical text-only backbone model.
  • User Evaluation: Won blind listening tests with human raters, confirming high perceptual quality and accuracy in reproducing difficult words.

Industry Insight

  • Enhanced Professional Usability: TTS systems can now be deployed in contexts requiring high precision for names, technical jargon, or loanwords without the need for costly manual phoneme labeling or speaker-specific fine-tuning.
  • Standardization of Pronunciation Control: The ability to inject pronunciation hints dynamically suggests a shift toward more modular TTS architectures where pronunciation and identity are decoupled, offering greater flexibility for developers.
  • Competitive Advantage: Systems integrating GRAFT-like capabilities will likely set a new standard for zero-shot TTS quality, particularly in multilingual and specialized domain applications, pushing competitors to adopt similar fine-grained conditioning techniques.

TL;DR

  • 提出GRAFT机制,通过植入参考音频片段解决零样本TTS中罕见专有名词和技术术语的发音歧义问题。
  • 利用模型自带的语音分词器编码参考音频,并将其绑定到提示词中的特定单词位置,实现细粒度的发音控制。
  • 在训练数据构建阶段引入声音转换技术,解耦提示说话人与目标说话人,确保参考音频可来自任意声音而不影响最终音色一致性。
  • 盲听测试显示人类评分者认为GRAFT生成的困难单词最接近参考录音;五语言基准测试将目标单词音素错误率降低22-39%。

为什么值得看

该研究突破了现有零样本TTS系统在处理非标准词汇时的发音瓶颈,为高精度语音合成提供了新的技术路径。对于需要生成包含大量专业术语或多语言混合内容的AI应用而言,这一方案显著提升了输出的准确性和自然度。

技术解析

  • 核心机制:GRAFT是一种针对神经编解码语言模型的逐词发音条件控制机制,旨在弥补文本条件甚至音素条件模型在直接声学控制上的不足。
  • 实现方式:系统从简短的口语样本中提取特定单词的发音,使用模型自身的语音分词器进行编码,并将该编码嵌入到提示序列中对应单词的位置。
  • 声音解耦策略:在训练数据准备过程中,采用声音转换技术将参考音频的说话人特征与目标说话人分离,使得参考发音来源多样化且不影响最终生成的目标音色。
  • 性能表现:在英语盲听研究中表现优异,在五语言客观基准测试中,相比相同的纯文本骨干网络,显著降低了目标单词的音素错误率,同时保持了说话人相似性和自然度。

行业启示

  • 精细化控制成为趋势:随着TTS技术成熟,行业竞争焦点正从整体自然度转向对特定词汇、情感或风格的细粒度精准控制能力。
  • 多模态融合解决长尾问题:结合声学参考信号与文本/音素输入是解决领域专有名词发音问题的有效手段,建议关注此类混合条件建模方法。
  • 数据工程的重要性:通过声音转换等预处理手段解耦不同声学特征,证明了高质量、结构化训练数据构建对提升模型泛化能力和鲁棒性的关键作用。

Disclaimer: The above content is generated by AI and is for reference only. 免责声明:以上内容由 AI 生成,仅供参考。

Speech 语音 Research 科学研究 Fine-tuning 微调