mlsanityData to 14 Oct 2026

PARCO: Phoneme-Augmented Robust Contextual ASR via Contrastive Entity Disambiguation

arXiv 2509.04357 ↗First posted 4 Sep 2025cs.CL cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Automatic speech recognition (ASR) systems struggle with domain-specific named entities, especially homophones. Contextual ASR improves recognition but often fails to capture fine-grained phoneme variations due to limited entity diversity. Moreover, prior methods treat entities as independent tokens, leading to incomplete multi-token biasing. To address these issues, we propose Phoneme-Augmented Robust Contextual ASR via COntrastive entity disambiguation (PARCO), which integrates phoneme-aware encoding, contrastive entity disambiguation, entity-level supervision, and hierarchical entity filtering. These components enhance phonetic discrimination, ensure complete entity retrieval, and reduce false positives under uncertainty. Experiments show that PARCO achieves CER of 4.22% on Chinese AISHELL-1 and WER of 11.14% on English DATA2 under 1,000 distractors, significantly outperforming baselines. PARCO also demonstrates robust gains on out-of-domain datasets like THCHS-30 and LibriSpeech.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 33 of its 46 references that are in the map. First posted on arXiv 4 Sep 2025.

Builds on

33 of 46 references.

References · 33

  1. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,863
  2. 2017AISHELL-1: An Open-Source Mandarin Speech Corpus and A Speech Recognition Baselinecited by 373
  3. 2016Joint CTC-Attention based End-to-End Speech Recognition using Multi-task Learningcited by 257
  4. 2018Deep context: end-to-end contextual speech recognitioncited by 137
  5. 2020Deep Shallow Fusion for RNN-T Personalizationcited by 64
  6. 2021Context-Aware Transformer Transducer for Speech Recognitioncited by 56
  7. 2024Automatic Speech Recognition using Advanced Deep Learning Approaches: A surveycited by 50
  8. 2015THCHS-30 : A Free Chinese Speech Corpuscited by 43
  9. 2022Improving End-to-End Contextual Speech Recognition with Fine-Grained Contextual Knowledge Selectioncited by 39
  10. 2020End-to-end Named Entity Recognition from English Speechcited by 36
  11. 2021Factorized Neural Transducer for Efficient Language Model Adaptationcited by 32
  12. 2021Instant One-Shot Word-Learning for Context-Specific Neural Sequence-to-Sequence Speech Recognitioncited by 30
21 more
  1. 2020CIF-based Collaborative Decoding for End-to-end Contextual Speech Recognitioncited by 21
  2. 2022AISHELL-NER: Named Entity Recognition from Chinese Speechcited by 19
  3. 2024Contextualized Automatic Speech Recognition with Attention-Based Bias Phrase Boosted Beam Searchcited by 19
  4. 2022Tree-constrained Pointer Generator with Graph Neural Network Encodings for Contextual Speech Recognitioncited by 17
  5. 2024Improving ASR Contextual Biasing with Guided Attentioncited by 17
  6. 2024Speech Emotion Recognition with ASR Transcripts: A Comprehensive Study on Word Error Rate and Fusion Techniquescited by 15
  7. 2023CopyNE: Better Contextual ASR by Copying Named Entitiescited by 14
  8. 2023Phoneme-aware Encoding for Prefix-tree-based Contextual ASRcited by 13
  9. 2024Improving Neural Biasing for Contextual Speech Recognition by Early Context Injection and Text Perturbationcited by 11
  10. 2023PROCTER: PROnunciation-aware ConTextual adaptER for personalized speech recognition in neural transducerscited by 10
  11. 2022Internal Language Model Estimation Through Explicit Context Vector Learning for Attention-based Encoder-decoder ASRcited by 9
  12. 2024Crossmodal ASR Error Correction with Discrete Speech Unitscited by 9
  13. 2024MF-AED-AEC: Speech Emotion Recognition by Leveraging Multimodal Fusion, Asr Error Detection, and Asr Error Correctioncited by 8
  14. 2022End-to-end contextual asr based on posterior distribution adaptation for hybrid ctc/attention systemcited by 7
  15. 2023ed-cec: improving rare word recognition using asr postprocessing based on error detection and context-aware error correctioncited by 7
  16. 2024Dynamic Encoder Size Based on Data-Driven Layer-wise Pruning for Speech Recognitioncited by 7
  17. 2023CIF-T: A Novel CIF-based Transducer Architecture for Automatic Speech Recognitioncited by 6
  18. 20242DP-2MRC: 2-Dimensional Pointer-based Machine Reading Comprehension Method for Multimodal Moment Retrievalcited by 2
  19. 2024Two-stage Framework for Robust Speech Emotion Recognition Using Target Speaker Extraction in Human Speech Noise Conditionscited by 2
  20. 2025GIA-MIC: Multimodal Emotion Recognition with Gated Interactive Attention and Modality-Invariant Learning Constraintscited by 1
  21. 2025PMF-CEC: Phoneme-augmented Multimodal Fusion for Context-aware ASR Error Correction with Error-specific Selective Decodingcited by 1

Built on by

Not cited yet.