English

BEST-STD2.0: Balanced and Efficient Speech Tokenizer for Spoken Term Detection

Audio and Speech Processing 2026-02-19 v2

Abstract

Fast and accurate spoken content retrieval is vital for applications such as voice search. Query-by-Example Spoken Term Detection (STD) involves retrieving matching segments from an audio database given a spoken query. Token-based STD systems, which use discrete speech representations, enable efficient search but struggle with robustness to noise and reverberation, and with inefficient token utilization. We address these challenges by proposing a noise and reverberation-augmented training strategy to improve tokenizer robustness. In addition, we introduce optimal transport-based regularization to ensure balanced token usage and enhance token efficiency. To further speed up retrieval, we adopt a TF-IDF-based search mechanism. Empirical evaluations demonstrate that the proposed method outperforms STD baselines across various distortion levels while maintaining high search efficiency.

Keywords

Cite

@article{arxiv.2512.16395,
  title  = {BEST-STD2.0: Balanced and Efficient Speech Tokenizer for Spoken Term Detection},
  author = {Anup Singh and Vipul Arora and Kris Demuynck},
  journal= {arXiv preprint arXiv:2512.16395},
  year   = {2026}
}

Comments

Accepted in ICASSP 2026

R2 v1 2026-07-01T08:31:05.428Z