1.2k citations · 1.8k across the 8 of their papers we have counts for
13 papers
E-Branchformer: Branchformer with Enhanced merging for speech recognition
Kwangyoun Kim, Felix Wu, Yifan Peng +4
Conformer, combining convolution and self-attention sequentially to capture both local and global information, has shown remarkable performance and is currently regarded as the sta…
Wav2Seq: Pre-training Speech-to-Text Encoder-Decoder Models Using Pseudo Languages
Felix Wu, Kwangyoun Kim, Shinji Watanabe +4
We introduce Wav2Seq, the first self-supervised approach to pre-train both parts of encoder-decoder models for speech data. We induce a pseudo language as a compact discrete repres…
Performance-Efficiency Trade-offs in Unsupervised Pre-training for Speech Recognition
Felix Wu, Kwangyoun Kim, Jing Pan +3
This paper is a study of performance-efficiency trade-offs in pre-trained models for automatic speech recognition (ASR). We focus on wav2vec 2.0, and formalize several architecture…
Revisiting Few-sample BERT Fine-tuning
Tianyi Zhang, Felix Wu, Arzoo Katiyar +2
This paper is a study of fine-tuning of BERT contextual representations, with focus on commonly observed instabilities in few-sample scenarios. We identify several factors that cau…
On Feature Normalization and Data Augmentation
Boyi Li, Felix Wu, Ser-Nam Lim +2
The moments (a.k.a., mean and standard deviation) of latent features are often removed as noise when training image recognition models, to increase stability and reduce training ti…
Positional Normalization
Boyi Li, Felix Wu, Kilian Q. Weinberger +1
A popular method to reduce the training time of deep neural networks is to normalize activations at each layer. Although various normalization schemes have been proposed, they all…