# Pretraining End-to-End Keyword Search with Automatically Discovered Acoustic Units

arXiv: https://arxiv.org/abs/2407.04652 · HTML: https://mlsanity.com/paper/2407.04652/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 35 references that are in the map.
- First posted on arXiv 5 Jul 2024.

## Builds on (most cited first, 12 of 12)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://mlsanity.com/paper/2111.09296/) (2021)
- [Libri-Light: A Benchmark for ASR with Limited or No Supervision](https://mlsanity.com/paper/1912.07875/) (2019)
- [vq-wav2vec: Self-Supervised Learning of Discrete Speech Representations](https://mlsanity.com/paper/1910.05453/) (2019)
- [Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech](https://mlsanity.com/paper/1911.09602/) (2019)
- [End-to-End ASR-free Keyword Search from Speech](https://mlsanity.com/paper/1701.04313/) (2017)
- [Bayesian Subspace Hidden Markov Model for Acoustic Unit Discovery](https://mlsanity.com/paper/1904.03876/) (2019)
- [A Hierarchical Subspace Model for Language-Attuned Acoustic Unit Discovery](https://mlsanity.com/paper/2011.03115/) (2020)
- [CNN-based Spoken Term Detection and Localization without Dynamic Programming](https://mlsanity.com/paper/2103.05468/) (2021)
- [End-to-End Open Vocabulary Keyword Search](https://mlsanity.com/paper/2108.10357/) (2021)
- [Spoken Term Detection and Relevance Score Estimation using Dot-Product of Pronunciation Embeddings](https://mlsanity.com/paper/2210.11895/) (2022)
- [Unsupervised Word Segmentation from Discrete Speech Units in Low-Resource Settings](https://mlsanity.com/paper/2106.04298/) (2021)

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

- [Vocal Tract Length Warped Features for Spoken Keyword Spotting](https://mlsanity.com/paper/2501.03523/) (2025)

## Abstract (arXiv)

End-to-end (E2E) keyword search (KWS) has emerged as an alternative and complimentary approach to conventional keyword search which depends on the output of automatic speech recognition (ASR) systems. While E2E methods greatly simplify the KWS pipeline, they generally have worse performance than their ASR-based counterparts, which can benefit from pretraining with untranscribed data. In this work, we propose a method for pretraining E2E KWS systems with untranscribed data, which involves using acoustic unit discovery (AUD) to obtain discrete units for untranscribed data and then learning to locate sequences of such units in the speech. We conduct experiments across languages and AUD systems: we show that finetuning such a model significantly outperforms a model trained from scratch, and the performance improvements are generally correlated with the quality of the AUD system used for pretraining.