The Continuous Integrate-and-Fire (CIF) mechanism provides effective alignment for non-autoregressive (NAR) speech recognition. This mechanism creates a smooth and monotonic mapping from acoustic features to target tokens, achieving performance on Mandarin competitive with other NAR approaches. However, without finer-grained guidance, its stability degrades in some languages such as English and French. In this paper, we propose Multi-scale CIF (M-CIF), which performs multi-level alignment by integrating character and phoneme level supervision progressively distilled into subword representations, thereby enhancing robust acoustic-text alignment. Experiments show that M-CIF reduces WER compared to the Paraformer baseline, especially on CommonVoice by 4.21% in German and 3.05% in French. To further investigate these gains, we define phonetic confusion errors (PE) and space-related segmentation errors (SE) as evaluation metrics. Analysis of these metrics across different M-CIF settings reveals that the phoneme and character layers are essential for enhancing progressive CIF alignment.
翻译:连续积分与触发(CIF)机制为非自回归(NAR)语音识别提供了有效的对齐方式。该机制通过建立声学特征与目标标记之间平滑且单调的映射,在普通话任务上取得了与其他NAR方法相竞争的性能。然而,由于缺乏细粒度指导,其在英语、法语等语言中的稳定性有所下降。本文提出多尺度CIF(M-CIF),通过整合逐步蒸馏到子词表示中的字符与音素层级监督,实现多层级对齐,从而增强声学-文本对齐的鲁棒性。实验表明,相较于Paraformer基线模型,M-CIF显著降低了词错误率(WER),尤其在CommonVoice数据集的德语和法语任务上分别降低了4.21%和3.05%。为深入探究性能提升原因,我们定义了音素混淆错误(PE)和空格相关分割错误(SE)作为评估指标。通过对不同M-CIF配置下这些指标的分析,发现音素层与字符层对于增强渐进式CIF对齐至关重要。