用可微分动态规划实现端到端语音音素对齐，性能超越现有方法。
Fully Differentiable Neural Forced Alignment via Soft Dynamic Programming
- 基于可微分软动态规划构建端到端对齐模型，支持全程梯度优化。
- 在英语标注数据上对齐准确率优于当前最佳方法，跨语言泛化能力强。
- 适合需要高精度语音对齐的语音识别与语音合成任务研究者使用。
序列建模的最新进展显著提升了自动语音识别（ASR）系统的性能，使其接近人类水平的识别准确率，并增强了在多种声学条件和语言下的鲁棒性。相比之下，强制对齐（Forced Alignment）并未取得相应进展，传统基于HMM-GMM的框架仍被广泛采用且具有很强竞争力。为弥补这一差距，我们提出一种专为音素对齐设计的端到端、完全可微分的神经架构。该模型包含一个处理输入信号的编码器和一个生成对齐决策的解码器。编码器由两个互补分支组成：一个用于音素身份验证，另一个用于音素边界检测。解码器采用基于可微分软动态规划的可训练模块。整个系统通过一种新型对比损失进行端到端优化，该损失鼓励稳态音素区域与过渡边界之间的清晰分离。所提方法在人工标注的英语基准测试中优于当前最先进的音素对齐技术，在词级别表现出良好的泛化能力，并在未见过的语言上也展现出良好泛化性能。
原文摘要 · Abstract (English)
Recent advances in sequence modeling have significantly improved ASR systems, bringing them close to human-level recognition accuracy and enhancing robustness across diverse acoustic conditions and languages. In contrast, Forced Alignment has not experienced comparable progress, and traditional HMM-GMM frameworks remain widely adopted and highly competitive. To address this gap, we propose an end-to-end, fully differentiable neural architecture specifically designed for phoneme alignment. The model consists of an encoder that processes the input signal and a decoder that produces alignment decisions. The encoder is structured into two complementary branches: one dedicated to phoneme identity verification and the other to phoneme boundary detection. The decoder is implemented as a trainable module based on differentiable soft dynamic programming. The entire system is optimized end-to-end using a novel contrastive loss that encourages clear separation between steady-state phoneme regions and transition boundaries. The proposed approach outperforms the current state of the art in phoneme alignment on hand-annotated English benchmarks, achieves strong word-level generalization results, and demonstrates generalization on unseen languages.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。