# The Pipeline System of ASR and NLU with MLM-based Data Augmentation toward STOP Low-resource Challenge

arXiv: https://arxiv.org/abs/2305.01194 · HTML: https://mlsanity.com/paper/2305.01194/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 6 references that are in the map.
- First posted on arXiv 2 May 2023.

## Builds on (most cited first, 5 of 5)

- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Low-Resource Domain Adaptation for Compositional Task-Oriented Semantic Parsing](https://mlsanity.com/paper/2010.03546/) (2020)
- [STOP: A dataset for Spoken Task Oriented Semantic Parsing](https://mlsanity.com/paper/2207.10643/) (2022)
- [Generate-and-Retrieve: use your predictions to improve retrieval for semantic parsing](https://mlsanity.com/paper/2209.14899/) (2022)
- [Generating Synthetic Data for Task-Oriented Semantic Parsing with Hierarchical Representations](https://mlsanity.com/paper/2011.02050/) (2020)

## Built on by (most cited first, 1 of 1)

- [Augmenting text for spoken language understanding with Large Language Models](https://mlsanity.com/paper/2309.09390/) (2023)

## Abstract (arXiv)

This paper describes our system for the low-resource domain adaptation track (Track 3) in Spoken Language Understanding Grand Challenge, which is a part of ICASSP Signal Processing Grand Challenge 2023. In the track, we adopt a pipeline approach of ASR and NLU. For ASR, we fine-tune Whisper for each domain with upsampling. For NLU, we fine-tune BART on all the Track3 data and then on low-resource domain data. We apply masked LM (MLM) -based data augmentation, where some of input tokens and corresponding target labels are replaced using MLM. We also apply a retrieval-based approach, where model input is augmented with similar training samples. As a result, we achieved exact match (EM) accuracy 63.3/75.0 (average: 69.15) for reminder/weather domain, and won the 1st place at the challenge.