paper-with-me

홈 › Papers

AdaMER-CTC: Connectionist Temporal Classification with Adaptive Maximum Entropy Regularization for Automatic Speech Recognition

2024-03-18 · SooHwan Eom, Eunseop Yoon, Hee Suk Yoon, Chanwoo Kim, Mark Hasegawa-Johnson, Chang D. Yoo

In Automatic Speech Recognition (ASR) systems, a recurring obstacle is the generation of narrowly focused output distributions. This phenomenon emerges as a side effect of Connectionist Temporal Classification (CTC), a robust sequence learning tool that utilizes dynamic programming for sequence mapping. While earlier efforts have tried to combine the CTC loss with an entropy maximization regularization term to mitigate this issue, they employed a constant weighting term on the regularization during the training, which we find may not be optimal. In this work, we introduce Adaptive Maximum Entropy Regularization (AdaMER), a technique that can modulate the impact of entropy regularization throughout the training process. This approach not only refines ASR model training but ensures that as training proceeds, predictions display the desired model confidence.

📄 PDF Abstract BibTeX arXiv:2403.11578

Code (0)

등록된 구현이 없습니다.

Tasks

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognitionSpeech Recognition

Methods 이 논문이 사용한 방법론

Entropy Regularization 설명 없음
CTC Loss 설명 없음

Similar Papers 제목 키워드 기반

AdaMerge: Salience-Aware Adaptive Token Merging for Training-Free Acceleration of Vision Transformers

2026-05-26 · Semi Lee, Hyejin Go, Hyesong Choi arxiv

The quadratic cost of self-attention in Vision Transformers (ViTs) constitutes a fundamental bottleneck for practical deployment, motivating a vibrant line of research on token reduction. Among existing approaches, token…

AdaMerging: Adaptive Model Merging for Multi-Task Learning

2023-10-04 · Enneng Yang, Zhenyi Wang, Li Shen, Shiwei Liu 외

Multi-task learning (MTL) aims to empower a model to tackle multiple tasks simultaneously. A recent development known as task arithmetic has revealed that several models, each fine-tuned for distinct tasks, can be direct…

modelMulti-Task LearningTask Arithmetic

Connectionist Temporal Classification with Maximum Entropy Regularization

2018-12-01 · NeurIPS 2018 12 · Hu Liu, Sheng Jin, Chang-Shui Zhang

Connectionist Temporal Classification (CTC) is an objective function for end-to-end sequence learning, which adopts dynamic programming algorithms to directly learn the mapping between sequences. CTC has shown promising …

ClassificationGeneral ClassificationScene Text Recognitionspeech-recognition+1

Improving CTC-AED model with integrated-CTC and auxiliary loss regularization

2023-08-15 · Daobin Zhu, Xiangdong Su, Hongbin Zhang

Connectionist temporal classification (CTC) and attention-based encoder decoder (AED) joint training has been widely applied in automatic speech recognition (ASR). Unlike most hybrid models that separately calculate the …

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Decoderspeech-recognition+1

Simultaneous Neural Machine Translation using Connectionist Temporal Classification

2019-11-27 · Katsuki Chousa, Katsuhito Sudoh, Satoshi Nakamura

Simultaneous machine translation is a variant of machine translation that starts the translation process before the end of an input. This task faces a trade-off between translation accuracy and latency. We have to determ…

ClassificationGeneral ClassificationMachine TranslationTranslation