# Cross-Lingual Query-by-Example Spoken Term Detection: A Transformer-Based Approach

arXiv: https://arxiv.org/abs/2410.04091 · HTML: https://mlsanity.com/paper/2410.04091/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 37 references that are in the map.
- First posted on arXiv 5 Oct 2024.

## Builds on (most cited first, 8 of 8)

- [Unsupervised Cross-lingual Representation Learning for Speech Recognition](https://mlsanity.com/paper/2006.13979/) (2020)
- [Deep convolutional acoustic word embeddings using word-pair side information](https://mlsanity.com/paper/1510.01032/) (2015)
- [Discriminative Acoustic Word Embeddings: Recurrent Neural Network-Based Approaches](https://mlsanity.com/paper/1611.02550/) (2016)
- [Query-by-Example Search with Discriminative Neural Acoustic Word Embeddings](https://mlsanity.com/paper/1706.03818/) (2017)
- [Neural Network based End-to-End Query by Example Spoken Term Detection](https://mlsanity.com/paper/1911.08332/) (2019)
- [Acoustic span embeddings for multilingual query-by-example search](https://mlsanity.com/paper/2011.11807/) (2020)
- [Query-by-example Spoken Term Detection using Attention-based Multi-hop Networks](https://mlsanity.com/paper/1709.00354/) (2017)
- [Multilingual Bottleneck Features for Query by Example Spoken Term Detection](https://mlsanity.com/paper/1907.00443/) (2019)

## Abstract (arXiv)

Query-by-example spoken term detection (QbE-STD) is typically constrained by transcribed data scarcity and language specificity. This paper introduces a novel, language-agnostic QbE-STD model leveraging image processing techniques and transformer architecture. By employing a pre-trained XLSR-53 network for feature extraction and a Hough transform for detection, our model effectively searches for user-defined spoken terms within any audio file. Experimental results across four languages demonstrate significant performance gains (19-54%) over a CNN-based baseline. While processing time is improved compared to DTW, accuracy remains inferior. Notably, our model offers the advantage of accurately counting query term repetitions within the target audio.