M-CIF: Multi-Scale Alignment For CIF-Based Non-Autoregressive ASR

arXiv:2510.22172 · cs.SD, cs.CL · Submitted 2025-10-25 · Read on arXiv

cs.SD, cs.CL

Submitted: 2025-10-25

Updated: 2026-09-21

Comments: Accepted by CCL 2026

Code: https://github.com/Moriiikdt/M-CIF

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Alignment — 49× in this paper · explained in 1 episode
• Transformer — 5× in this paper · explained in 2 episodes
• Ablation — 4× in this paper · explained in 1 episode
• Accuracy — 3× in this paper · explained in 1 episode
• Continuous Integrate-and-Fire — 3× in this paper
• Fidelity — 3× in this paper · explained in 2 episodes
• Iterative Refinement — 2× in this paper · explained in 3 episodes
• Ablation study — 1× in this paper · explained in 1 episode
• Data Augmentation — 1× in this paper · explained in 7 episodes
• Interpretability — 1× in this paper · explained in 15 episodes
• Regularization — 1× in this paper · explained in 3 episodes
• Tokenization — 1× in this paper · explained in 2 episodes
• Word Error Rate — 1× in this paper

Sources

• Paraformer: Fast and Accurate Parallel Transformer for Non-autoregressive End-to-End Speech Recognition
• SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing
• Conformer: Convolution-augmented Transformer for Speech Recognition
• An efficient text augmentation approach for contextualized Mandarin speech recognition
• Common Voice: A Massively-Multilingual Speech Corpus
• Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinement
• Soundwave: Less is More for Speech-Text Alignment in LLMs
• FunASR: A Fundamental End-to-End Speech Recognition Toolkit
• SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition

Related papers

• Few-Shot Open-Set Audio Classification via Transductive Prototype Refinement and Class Logit Enhancement
• Spectral Masking and Interpolation Attack (SMIA): A Black-box Adversarial Attack against Voice Authentication and Anti-Spoofing Systems
• AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and Thinking
• SoundWeaver: Compositional Warm-Starting for Text-to-Audio Diffusion Serving
• WASIL: In-the-Wild Arabic Spoken Interactions with LLMs
• Efficient Audiovisual Speech Processing via MUTUD: Multimodal Training and Unimodal Deployment