paper-with-me

홈 › Papers

Contrastive Semi-supervised Learning for ASR

2021-03-09 · Alex Xiao, Christian Fuegen, Abdelrahman Mohamed

Pseudo-labeling is the most adopted method for pre-training automatic speech recognition (ASR) models. However, its performance suffers from the supervised teacher model's degrading quality in low-resource setups and under domain transfer. Inspired by the successes of contrastive representation learning for computer vision and speech applications, and more recently for supervised learning of visual objects, we propose Contrastive Semi-supervised Learning (CSL). CSL eschews directly predicting teacher-generated pseudo-labels in favor of utilizing them to select positive and negative examples. In the challenging task of transcribing public social media videos, using CSL reduces the WER by 8% compared to the standard Cross-Entropy pseudo-labeling (CE-PL) when 10hr of supervised data is used to annotate 75,000hr of videos. The WER reduction jumps to 19% under the ultra low-resource condition of using 1hr labels for teacher supervision. CSL generalizes much better in out-of-domain conditions, showing up to 17% WER reduction compared to the best CE-PL pre-trained model.

📄 PDF Abstract BibTeX arXiv:2103.05149

Code (0)

등록된 구현이 없습니다.

Tasks

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Representation Learningspeech-recognitionSpeech Recognition

Methods 이 논문이 사용한 방법론

CSL Circular Smooth Label (CSL) is a classification-based rotation detection technique for arbitrary-oriented object detection. It is used for circularly distributed angle…

Similar Papers 제목 키워드 기반

Semi-Supervised Relational Contrastive Learning

2023-04-11 · Attiano Purpura-Pontoniere, Demetri Terzopoulos, Adam Wang, Abdullah-Al-Zubaer Imran

Disease diagnosis from medical images via supervised learning is usually dependent on tedious, error-prone, and costly image labeling by medical experts. Alternatively, semi-supervised learning and self-supervised learni…

Contrastive LearningDiagnosticLesion ClassificationSelf-Supervised Learning+1

CoDiM: Learning with Noisy Labels via Contrastive Semi-Supervised Learning

2021-11-23 · Xin Zhang, Zixuan Liu, Kaiwen Xiao, Tian Shen 외

Labels are costly and sometimes unreliable. Noisy label learning, semi-supervised learning, and contrastive learning are three different strategies for designing learning processes requiring less annotation cost. Semi-su…

Contrastive LearningImage ClassificationLearning with noisy labels

Rethinking Weak Supervision in Helping Contrastive Learning

2023-06-07 · Jingyi Cui, Weiran Huang, Yifei Wang, Yisen Wang

Contrastive learning has shown outstanding performances in both supervised and unsupervised learning, and has recently been introduced to solve weakly supervised learning problems such as semi-supervised learning and noi…

Contrastive LearningDenoisingWeakly-supervised Learning

Region-level Contrastive and Consistency Learning for Semi-Supervised Semantic Segmentation

2022-04-28 · Jianrong Zhang, Tianyi Wu, Chuanghao Ding, Hongwei Zhao 외

Current semi-supervised semantic segmentation methods mainly focus on designing pixel-level consistency and contrastive regularization. However, pixel-level regularization is sensitive to noise from pixels with incorrect…

SegmentationSemantic SegmentationSemi-Supervised Semantic Segmentation

Semi-supervised Contrastive Learning with Similarity Co-calibration

2021-05-16 · Yuhang Zhang, Xiaopeng Zhang, Robert. C. Qiu, Jie Li 외

Semi-supervised learning acts as an effective way to leverage massive unlabeled data. In this paper, we propose a novel training strategy, termed as Semi-supervised Contrastive Learning (SsCL), which combines the well-kn…

Contrastive LearningFew-Shot LearningSelf-Supervised Learning