动态自动语音识别路径：一种自适应掩码方法，实现多语言自动语音识别模型的高效剪枝。

摘要

神经网络剪枝提供了一种有效的方法，可以在最小性能损失的情况下压缩多语言自动语音识别（ASR）模型。然而，这需要多轮剪枝和重新训练，每种语言都需要运行。在这项工作中，我们提出了在两种场景下为了高效剪枝多语言ASR模型而使用自适应掩模方法，分别导致稀疏的单语模型或稀疏的多语言模型（称为动态ASR路径）。我们的方法动态地调整子网络，避免对固定子网络结构做出过早决定。我们展示了我们的方法在针对稀疏的单语模型时优于现有的剪枝方法。此外，我们阐明了动态ASR路径共同发现并训练了更好的单个多语言模型的子网络（路径），通过从不同的子网络初始化进行调整，从而减少了对特定语言剪枝的需求。

English

Neural network pruning offers an effective method for compressing a multilingual automatic speech recognition (ASR) model with minimal performance loss. However, it entails several rounds of pruning and re-training needed to be run for each language. In this work, we propose the use of an adaptive masking approach in two scenarios for pruning a multilingual ASR model efficiently, each resulting in sparse monolingual models or a sparse multilingual model (named as Dynamic ASR Pathways). Our approach dynamically adapts the sub-network, avoiding premature decisions about a fixed sub-network structure. We show that our approach outperforms existing pruning methods when targeting sparse monolingual models. Further, we illustrate that Dynamic ASR Pathways jointly discovers and trains better sub-networks (pathways) of a single multilingual model by adapting from different sub-network initializations, thereby reducing the need for language-specific pruning.

动态自动语音识别路径：一种自适应掩码方法，实现多语言自动语音识别模型的高效剪枝。

Dynamic ASR Pathways: An Adaptive Masking Approach Towards Efficient Pruning of A Multilingual ASR Model

摘要

Support