11 accepted papers
Automatic speech recognition (ASR) systems can benefit from incorporating contextual information to improve recognition accuracy, especially for uncommon words or phrases. Current approaches like custom vocabularies or prompting with previous transcript segments provide limited contextual control. C…
This work proposes a new parameter-efficient adaptation approach for multilingual ASR based on the hyper-network. Existing multilingual ASR adaptation methods apply either one residual adapter for all the languages, or language dependent adapters for each individual language. The residual adapter ca…
Self-attention based transformer models have been dominating many computer vision tasks in the past few years. Their superb model qualities heavily depend on the excessively large labeled image datasets. In order to reduce the reliance on large labeled datasets, reconstruction based masked autoencod…
Despite recent advancements in speech processing, zero-resource speech translation (ST) and automatic speech recognition (ASR) remain challenging problems. In this work, we propose to leverage a multilingual Large Language Model (LLM) to perform ST and ASR in languages for which the model has never…
Integrated Speech and Large Language Models (SLMs) that can follow speech instructions and generate relevant text responses have gained popularity lately. However, the safety and robustness of these models remains largely unclear. In this work, we investigate the potential vulnerabilities of such in…
Channel pruning has been broadly recognized as an effective technique to reduce the computation and memory cost of deep convolutional neural networks. However, conventional pruning methods have limitations in that: they are restricted to pruning process only, and they require a fully pre-trained lar…
Deep neural networks (DNNs) are effective in solving many real-world problems. Larger DNN models usually exhibit better quality (e.g., accuracy) but their excessive computation results in long inference time. Model sparsification can reduce the computation and memory cost while maintaining model qua…
We consider few-shot learning (FSL), where a model learns from very few labeled examples such that it can generalize to unseen examples. Model-agnostic meta-learning (MAML) has been proposed to solve FSL. However, the low performance of MAML suggests its difficulty in tackle diverse tasks, due to th…
Deep convolutional neural networks (CNN) have demonstrated superior performance in image super-resolution (SR) problem. However, CNNs are known to be heavily over-parameterized, and suffer from abundant redundancy. The growing size of CNNs may be incompatible with their deployment on mobile or embed…
Kernel approximation methods create explicit, low-dimensional kernel feature maps to deal with the high computational and memory complexity of standard techniques. This work studies a supervised kernel learning methodology to optimize such mappings. We utilize the Discriminant Information criterion,…
Back-propagation (BP) is now a classic learning paradigm whose source of supervision is exclusively from the external (input/output) nodes. Consequently, BP is easily vulnerable to curse-of-depth in (very) Deep Learning Networks (DLNs). This prompts us to advocate Internal Neuron’s Learnablility (IN…