ASR Model

Automatic speech recognition (ASR) models aim to accurately transcribe spoken language into text, a task crucial for numerous applications. Current research emphasizes improving model robustness across diverse accents, languages, and noisy environments, often leveraging transformer-based architectures like Wav2Vec 2.0 and Conformer, and incorporating visual information for improved accuracy. Significant efforts focus on addressing biases in ASR models, enhancing efficiency through knowledge distillation and self-supervised learning, and developing methods for low-resource languages. These advancements are driving progress in various fields, including accessibility technologies, human-computer interaction, and language documentation.

Papers

January 26, 2022

Internal Language Model Estimation Through Explicit Context Vector Learning for Attention-based Encoder-decoder ASR
Yufei Liu, Rao Ma, Haihua Xu, Yi He, Zejun Ma, Weibin Zhang
Speech Recognition ASR Model Context Vector Internal Language Model Attention Based Encoder

November 5, 2021

Sequential Randomized Smoothing for Adversarially Robust Speech Recognition
Raphael Olivier, Bhiksha Raj
Adversarial Attack Automatic Speech Recognition Speech Recognition Randomized Smoothing ASR Model Adaptive Attack Imperceptible Attack

ASR Model

Papers

Internal Language Model Estimation Through Explicit Context Vector Learning for Attention-based Encoder-decoder ASR

Sequential Randomized Smoothing for Adversarially Robust Speech Recognition