تقرير
G-Augment: Searching for the Meta-Structure of Data Augmentation Policies for ASR
العنوان: | G-Augment: Searching for the Meta-Structure of Data Augmentation Policies for ASR |
---|---|
المؤلفون: | Wang, Gary, Cubuk, Ekin D., Rosenberg, Andrew, Cheng, Shuyang, Weiss, Ron J., Ramabhadran, Bhuvana, Moreno, Pedro J., Le, Quoc V., Park, Daniel S. |
سنة النشر: | 2022 |
المجموعة: | Computer Science |
مصطلحات موضوعية: | Computer Science - Machine Learning, Computer Science - Computation and Language, Computer Science - Sound, Electrical Engineering and Systems Science - Audio and Speech Processing |
الوصف: | Data augmentation is a ubiquitous technique used to provide robustness to automatic speech recognition (ASR) training. However, even as so much of the ASR training process has become automated and more "end-to-end", the data augmentation policy (what augmentation functions to use, and how to apply them) remains hand-crafted. We present Graph-Augment, a technique to define the augmentation space as directed acyclic graphs (DAGs) and search over this space to optimize the augmentation policy itself. We show that given the same computational budget, policies produced by G-Augment are able to perform better than SpecAugment policies obtained by random search on fine-tuning tasks on CHiME-6 and AMI. G-Augment is also able to establish a new state-of-the-art ASR performance on the CHiME-6 evaluation set (30.7% WER). We further demonstrate that G-Augment policies show better transfer properties across warm-start to cold-start training and model size compared to random-searched SpecAugment policies. Comment: 6 pages, accepted at SLT 2022. Updated with copyright |
نوع الوثيقة: | Working Paper |
URL الوصول: | http://arxiv.org/abs/2210.10879 |
رقم الأكسشن: | edsarx.2210.10879 |
قاعدة البيانات: | arXiv |
الوصف غير متاح. |