Locality enhanced dynamic biasing and sampling strategies for contextual ASR
Not on the map. Why?
Abstract (arXiv)
Automatic Speech Recognition (ASR) still face challenges when recognizing time-variant rare-phrases. Contextual biasing (CB) modules bias ASR model towards such contextually-relevant phrases. During training, a list of biasing phrases are selected from a large pool of phrases following a sampling strategy. In this work we firstly analyse different sampling strategies to provide insights into the training of CB for ASR with correlation plots between the bias embeddings among various training stages. Secondly, we introduce a neighbourhood attention (NA) that localizes self attention (SA) to the nearest neighbouring frames to further refine the CB output. The results show that this proposed approach provides on average a 25.84% relative WER improvement on LibriSpeech sets and rare-word evaluation compared to the baseline.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 31 of its 39 references that are in the map. First posted on arXiv 23 Jan 2024.
Builds on
31 of 39 references.
References · 31
- 2017Attention Is All You Needcited by 60,361
- 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
- 2020Training data-efficient image transformers & distillation through attentioncited by 4,586
- 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,863
- 2019Transformer-XL: Attentive Language Models Beyond a Fixed-Length Contextcited by 1,850
- 2016Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translationcited by 1,475
- 2019Stand-Alone Self-Attention in Vision Modelscited by 542
- 2021SpeechBrain: A General-Purpose Speech Toolkitcited by 406
- 2020LUKE: Deep Contextualized Entity Representations with Entity-aware Self-attentioncited by 326
- 2022Neighborhood Attention Transformercited by 239
- 2018Deep context: end-to-end contextual speech recognitioncited by 137
- 2021Contextualized Streaming End-to-End Speech Recognition with Trie-Based Deep Biasing and Shallow Fusioncited by 101
19 more
- 2020Contextual RNN-T For Open Domain ASRcited by 85
- 2022Contextual Adapters for Personalized Speech Recognition in Neural Transducerscited by 57
- 2021Context-Aware Transformer Transducer for Speech Recognitioncited by 56
- 2020Class LM and word mapping for contextual biasing in End-to-End ASRcited by 42
- 2019Personalization of End-to-end Speech Recognition On Mobile Devices For Named Entitiescited by 41
- 2022Improving End-to-End Contextual Speech Recognition with Fine-Grained Contextual Knowledge Selectioncited by 39
- 2018Contextual Speech Recognition with Difficult Negative Training Examplescited by 28
- 2021Fast Contextual Adaptation with Neural Associative Memory for On-Device Personalized Speech Recognitioncited by 25
- 2022Improving Contextual Recognition of Rare Words with an Alternate Spelling Prediction Modelcited by 17
- 2021A Light-weight contextual spelling correction model for customizing transducer-based speech recognition systemscited by 11
- 2022Minimising Biasing Word Errors for Contextual ASR with the Tree-Constrained Pointer Generatorcited by 11
- 2023CB-Conformer: Contextual biasing Conformer for biased word recognitioncited by 11
- 2023PROCTER: PROnunciation-aware ConTextual adaptER for personalized speech recognition in neural transducerscited by 10
- 2023Improving Contextual Spelling Correction by External Acoustics Attention and Semantic Aware Data Augmentationcited by 9
- 2023Approximate Nearest Neighbour Phrase Mining for Contextual Speech Recognitioncited by 8
- 2020Fast and Robust Unsupervised Contextual Biasing for Speech Recognitioncited by 7
- 2022End-to-end contextual asr based on posterior distribution adaptation for hybrid ctc/attention systemcited by 7
- 2023Dialog act guided contextual adapter for personalized speech recognitioncited by 4
- 2023Matching-based Term Semantics Pre-training for Spoken Patient Query Understandingcited by 2
Built on by
Citing papers per year: 2024: 1; 2025: 2.
3 papers.
- 2024Optimizing Contextual Speech Recognition Using Vector Quantization for Efficient Retrievalcited by 3
- 2025Enhancing the Robustness of Contextual ASR to Varying Biasing Information Volumes Through Purified Semantic Correlation Joint Modelingcited by 1
- 2025Retrieval Augmented Generation based context discovery for ASRcited by 1
- 2025Retrieval Augmented Generation based context discovery for ASR23 Sep 2025
- 2025Enhancing the Robustness of Contextual ASR to Varying Biasing Information Volumes Through Purified Semantic Correlation Joint Modeling7 Sep 2025
- 2024Optimizing Contextual Speech Recognition Using Vector Quantization for Efficient Retrieval1 Nov 2024