Large scale weakly and semi-supervised learning for low-resource video ASR

By Kritika Singh, Vimal Manohar, Alex Xiao, Sergey Edunov, Ross B. Girshick, Vitaliy Liptchinsky, Christian Fuegen, Yatharth Saraf, Geoffry Zweig, Abdel-rahman Mohamed. 2020

10 citations recorded by the source corpus.

Many semi- and weakly-supervised approaches have been investigated for overcoming the labeling cost of building high quality speech recognition systems. On the challenging task of transcribing social media videos in low-resource conditions, we conduct a large scale systematic comparison between two self-labeling methods on one hand, and weakly-supervised pretraining using contextual metadata on the other. We investigate distillation methods at the frame level and the sequence level for hybrid, encoder-only CTC-based, and encoder-decoder speech recognition systems on Dutch and Romanian languages using 27,000 and 58,000 hours of unlabeled audio respectively. Although all approaches improved upon their respective baseline WERs by more than 8%, sequence-level distillation for encoder-decoder models provided the largest relative WER reduction of 20% compared to the strongest data-augmented supervised baseline.

Subfields: Artificial Intelligence, Representation & Self-Supervised Learning, Speech & Audio, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Representation & Self-Supervised Learning

In Speech & Audio

In Signal Processing

Citations in the corpus

The People's Speech: A Large-Scale Diverse English Speech Recognition Dataset for Commercial Usage (2021, methodology): “If hypothesis (3) is true, we could use weakly supervised training methods on this unlabeled data, like [29].”
Contrastive Semi-Supervised Learning for ASR (2021, methodology): “Using multiple rounds of pseudo-labeling for model pre-training was found very effective for the classical GMM/HMM systems in the mid 2000s [5] and more recently for neural systems [11, 32, 8].”