# Phoneme-Level Contrastive Learning for User-Defined Keyword Spotting with Flexible Enrollment

arXiv: https://arxiv.org/abs/2412.20805 · HTML: https://mlsanity.com/paper/2412.20805/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Dec 2024.

## Built on by (most cited first, 5 of 5)

- [Identifying and typifying demographic unfairness in phoneme-level embeddings of self-supervised speech recognition models](https://mlsanity.com/paper/2604.22631/) (2026)
- [Cross-Lingual IPA Contrastive Learning for Zero-Shot NER](https://mlsanity.com/paper/2503.07214/) (2025)
- [Adversarial Deep Metric Learning for Cross-Modal Audio-Text Alignment in Open-Vocabulary Keyword Spotting](https://mlsanity.com/paper/2505.16735/) (2025)
- [MATE: Matryoshka Audio-Text Embeddings for Open-Vocabulary Keyword Spotting](https://mlsanity.com/paper/2601.14012/) (2026)
- [ProKWS: Personalized Keyword Spotting via Collaborative Learning of Phonemes and Prosody](https://mlsanity.com/paper/2603.18024/) (2026)

## Abstract (arXiv)

User-defined keyword spotting (KWS) enhances the user experience by allowing individuals to customize keywords. However, in open-vocabulary scenarios, most existing methods commonly suffer from high false alarm rates with confusable words and are limited to either audio-only or text-only enrollment. Therefore, in this paper, we first explore the model's robustness against confusable words. Specifically, we propose Phoneme-Level Contrastive Learning (PLCL), which refines and aligns query and source feature representations at the phoneme level. This method enhances the model's disambiguation capability through fine-grained positive and negative comparisons for more accurate alignment, and it is generalizable to jointly optimize both audio-text and audio-audio matching, adapting to various enrollment modes. Furthermore, we maintain a context-agnostic phoneme memory bank to construct confusable negatives for data augmentation. Based on this, a third-category discriminator is specifically designed to distinguish hard negatives. Overall, we develop a robust and flexible KWS system, supporting different modality enrollment methods within a unified framework. Verified on the LibriPhrase dataset, the proposed approach achieves state-of-the-art performance.