mlsanityData to 14 Oct 2026

Contextualized Token Discrimination for Speech Search Query Correction

arXiv 2509.04393 ↗First posted 4 Sep 2025cs.SD cs.CLcited by 1

Not on the map. Why?

Abstract (arXiv)

Query spelling correction is an important function of modern search engines since it effectively helps users express their intentions clearly. With the growing popularity of speech search driven by Automated Speech Recognition (ASR) systems, this paper introduces a novel method named Contextualized Token Discrimination (CTD) to conduct effective speech query correction. In CTD, we first employ BERT to generate token-level contextualized representations and then construct a composition layer to enhance semantic information. Finally, we produce the correct query according to the aggregated token representation, correcting the incorrect tokens by comparing the original token representations and the contextualized representations. Extensive experiments demonstrate the superior performance of our proposed method across all metrics, and we further present a new benchmark dataset with erroneous ASR transcriptions to offer comprehensive evaluations for audio query correction.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 34 of its 62 references that are in the map. First posted on arXiv 4 Sep 2025.

Builds on

34 of 62 references.

References · 34

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,375
  4. 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,905
  5. 2013Distributed Representations of Words and Phrases and their Compositionalitycited by 3,474
  6. 2018Deep contextualized word representationscited by 2,290
  7. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,863
  8. 2017Accurate, Large Minibatch SGD: Training ImageNet in 1 Hourcited by 1,492
  9. 2019How Contextual are Contextualized Word Representations? Comparing the Geometry of BERT, ELMo, and GPT-2 Embeddingscited by 784
  10. 2017AISHELL-1: An Open-Source Mandarin Speech Corpus and A Speech Recognition Baselinecited by 373
  11. 2016Enhanced LSTM for Natural Language Inferencecited by 200
  12. 2020Pushing the Limits of Semi-Supervised Learning for Automatic Speech Recognitioncited by 187
22 more
  1. 2018Deep context: end-to-end contextual speech recognitioncited by 137
  2. 2019End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architecturescited by 120
  3. 2018Exploring Architectures, Data and Units For Streaming End-to-End Speech Recognition with RNN-Transducercited by 96
  4. 2020Spelling Error Correction with Soft-Masked BERTcited by 90
  5. 2022The Past Mistake is the Future Wisdom: Error-driven Contrastive Probability Optimization for Chinese Spell Checkingcited by 56
  6. 2020ASR Error Correction and Domain Adaptation Using Machine Translationcited by 48
  7. 2020Table Search Using a Deep Contextualized Language Modelcited by 22
  8. 2020ESPnet-se: end-to-end speech enhancement and separation toolkit designed for asr integrationcited by 22
  9. 2023Rethinking Masked Language Modeling for Chinese Spelling Correctioncited by 19
  10. 2020Contextualized Embeddings in Named-Entity Recognition: An Empirical Study on Generalizationcited by 15
  11. 2023Disentangled Phonetic Representation for Chinese Spelling Correctioncited by 14
  12. 2024Dial-In LLM: Human-Aligned LLM-in-the-loop Intent Clustering for Customer Service Dialoguescited by 11
  13. 2021Content Selection Network for Document-grounded Retrieval-based Chatbotscited by 10
  14. 2018Familia: A Configurable Topic Modeling Framework for Industrial Text Engineeringcited by 7
  15. 2024ASR-EC Benchmark: Evaluating Large Language Models on Chinese ASR Error Correctioncited by 7
  16. 2025Contextual ASR Error Handling with LLMs Augmentation for Goal-Oriented Conversational AIcited by 6
  17. 2025QualBench: Benchmarking Chinese LLMs with Localized Professional Qualifications for Vertical Domain Evaluationcited by 5
  18. 2017Audio-to-score alignment of piano music using RNN-based automatic music transcriptioncited by 4
  19. 2021You Get What You Chat: Using Conversations to Personalize Search-based Recommendationscited by 4
  20. 2024InfantCryNet: A Data-driven Framework for Intelligent Analysis of Infant Criescited by 4
  21. 2024Robust ASR Error Correction with Conservative Data Filteringcited by 3
  22. 2024Acoustic Model Optimization over Multiple Data Sources: Merging and Valuationcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Federated Heterogeneous Language Model Optimization for Hybrid Automatic Speech Recognitioncited by 1