Adapting Pretrained Transformer to Lattices for Spoken Language Understanding
This work addresses spoken language understanding by improving accuracy for applications like speech recognition, though it is incremental as it adapts existing methods to a specific input type.
The paper tackled the problem of adapting pretrained transformers to encode lattices for spoken language understanding, showing that fine-tuning with lattice inputs yields clear improvement over using 1-best results on the ATIS dataset.
Lattices are compact representations that encode multiple hypotheses, such as speech recognition results or different word segmentations. It is shown that encoding lattices as opposed to 1-best results generated by automatic speech recognizer (ASR) boosts the performance of spoken language understanding (SLU). Recently, pretrained language models with the transformer architecture have achieved the state-of-the-art results on natural language understanding, but their ability of encoding lattices has not been explored. Therefore, this paper aims at adapting pretrained transformers to lattice inputs in order to perform understanding tasks specifically for spoken language. Our experiments on the benchmark ATIS dataset show that fine-tuning pretrained transformers with lattice inputs yields clear improvement over fine-tuning with 1-best results. Further evaluation demonstrates the effectiveness of our methods under different acoustic conditions. Our code is available at https://github.com/MiuLab/Lattice-SLU