M-CIF: Multi-Scale Alignment For CIF-Based Non-Autoregressive ASR
Mao, Ruixiang, Ma, Xiangnan, Yang, Qing, Zhu, Ziming, Qiao, Yucheng, Ge, Yuan, Xiao, Tong, Gao, Shengxiang, Yu, Zhengtao, Zhu, Jingbo
–arXiv.org Artificial Intelligence
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.
arXiv.org Artificial Intelligence
Oct-28-2025
- Country:
- Asia > China
- Beijing > Beijing (0.04)
- Liaoning Province > Shenyang (0.04)
- Yunnan Province > Kunming (0.04)
- North America > United States
- Illinois > Cook County
- Chicago (0.04)
- Pennsylvania > Allegheny County
- Pittsburgh (0.04)
- Illinois > Cook County
- South America > Chile
- Asia > China
- Genre:
- Research Report (0.50)
- Technology:
- Information Technology > Artificial Intelligence
- Machine Learning > Neural Networks (0.46)
- Natural Language (0.94)
- Speech > Speech Recognition (0.38)
- Information Technology > Artificial Intelligence