新对齐算法提升语音识别误差分析精度
A Text-To-Text Alignment Algorithm for Better Evaluation of Modern Speech Recognition Systems
- 结合动态规划与束搜索评分实现精准文本对齐
- 显著提升罕见词、专有名词等关键错误的对齐准确率
- 适合需要细粒度评估的语音识别研究与应用
现代神经网络在语音识别基准上表现大幅提升，但性能增长常由高频词驱动，其语义权重较低，导致词错误率（WER）等整体指标掩盖了真实差异。罕见词、命名实体及领域专有词汇的错误更具影响，却常被聚合指标隐藏。这凸显了精细化错误分析的需求，而其依赖于参考文本与模型输出间的精确对齐。现有对齐方法难以满足此要求。本文提出一种新型对齐算法，将动态规划与束搜索评分相结合，相比传统方法能更准确对齐个体错误，实现可靠错误分析。该算法已通过 PyPI 发布。
原文摘要 · Abstract (English)
Modern neural networks have greatly improved performance across speech recognition benchmarks. However, gains are often driven by frequent words with limited semantic weight, which can obscure meaningful differences in word error rate, the primary evaluation metric. Errors in rare terms, named entities, and domain-specific vocabulary are more consequential, but remain hidden by aggregate metrics. This highlights the need for finer-grained error analysis, which depends on accurate alignment between reference and model transcripts. However, conventional alignment methods are not designed for such precision. We propose a novel alignment algorithm that couples dynamic programming with beam search scoring. Compared to traditional text alignment methods, our approach provides more accurate alignment of individual errors, enabling reliable error analysis. The algorithm is made available via PyPI.
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。