mlsanityData to 14 Oct 2026

SeACo-Paraformer: A Non-Autoregressive ASR System with Flexible and Effective Hotword Customization Ability

arXiv 2308.03266 ↗First posted 7 Aug 2023cs.SD cs.CL eess.AScited by 18

Not on the map. Why?

Abstract (arXiv)

Hotword customization is one of the concerned issues remained in ASR field - it is of value to enable users of ASR systems to customize names of entities, persons and other phrases to obtain better experience. The past few years have seen effective modeling strategies for ASR contextualization developed, but they still exhibit space for improvement about training stability and the invisible activation process. In this paper we propose Semantic-Augmented Contextual-Paraformer (SeACo-Paraformer) a novel NAR based ASR system with flexible and effective hotword customization ability. It possesses the advantages of AED-based model's accuracy, NAR model's efficiency, and explicit customization capacity of superior performance. Through extensive experiments with 50,000 hours of industrial big data, our proposed model outperforms strong baselines in customization. Besides, we explore an efficient way to filter large-scale incoming hotwords for further improvement. The industrial models compared, source codes and two hotword test sets are all open source.

Key facts. Cited by 18 papers in the map (data through 14 Oct 2026). Builds on 22 of its 25 references that are in the map. First posted on arXiv 7 Aug 2023.

Builds on

22 of 25 references.

References · 22

  1. 2017Attention Is All You Needcited by 60,361
  2. 2013Speech Recognition with Deep Recurrent Neural Networkscited by 669
  3. 2015Listen, Attend and Spellcited by 543
  4. 2017AISHELL-1: An Open-Source Mandarin Speech Corpus and A Speech Recognition Baselinecited by 373
  5. 2019A Comparative Study on Transformer vs RNN in Speech Applicationscited by 211
  6. 2022Paraformer: Fast and Accurate Parallel Transformer for Non-autoregressive End-to-End Speech Recognitioncited by 144
  7. 2018Deep context: end-to-end contextual speech recognitioncited by 137
  8. 2018Exploring Architectures, Data and Units For Streaming End-to-End Speech Recognition with RNN-Transducercited by 96
  9. 2023FunASR: A Fundamental End-to-End Speech Recognition Toolkitcited by 95
  10. 2020Contextual RNN-T For Open Domain ASRcited by 85
  11. 2019CIF: Continuous Integrate-and-Fire for End-to-End Speech Recognitioncited by 77
  12. 2020Streaming automatic speech recognition with the transformer modelcited by 73
10 more
  1. 2022Contextual Adapters for Personalized Speech Recognition in Neural Transducerscited by 57
  2. 2022Improving End-to-End Contextual Speech Recognition with Fine-Grained Contextual Knowledge Selectioncited by 39
  3. 2023Contextualized End-to-End Speech Recognition with Contextual Phrase Prediction Networkcited by 31
  4. 2018Multilingual End-to-End Speech Recognition with A Single Transformer on Low-Resource Languagescited by 26
  5. 2020CIF-based Collaborative Decoding for End-to-end Contextual Speech Recognitioncited by 21
  6. 2020SAN-M: Memory Equipped Self-Attention for End-to-End Speech Recognitioncited by 16
  7. 2021Boundary and Context Aware Training for CIF-based Non-Autoregressive End-to-end ASRcited by 12
  8. 2018Chinese Lexical Analysis with Deep Bi-GRU-CRF Networkcited by 9
  9. 2020Universal ASR: Unifying Streaming and Non-Streaming ASR Using a Single Encoder-Decoder Modelcited by 9
  10. 2019Phoneme-Based Contextualization for Cross-Lingual Speech Recognition in End-to-End Modelscited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 5; 2025: 6; 2026: 6.

18 papers.

Sort:

By area:

  1. 2024FunAudioLLM: Voice Understanding and Generation Foundation Models for Natural Interaction Between Humans and LLMscited by 122
  2. 2024CTC-Assisted LLM-Based Contextual ASRcited by 19
  3. 2023A Multitask Training Approach to Enhance Whisper with Contextual Biasing and Open-Vocabulary Keyword Spottingcited by 5
  4. 2025Locate-and-Focus: Enhancing Terminology Translation in Speech Language Modelscited by 3
  5. 2025Zero-shot Context Biasing with Trie-based Decoding using Synthetic Multi-Pronunciationcited by 3
  6. 2024An efficient text augmentation approach for contextualized Mandarin speech recognitioncited by 2
  7. 2024VHASR: A Multimodal Speech Recognition System With Vision Hotwordscited by 2
  8. 2024XCB: an effective contextual biasing approach to bias cross-lingual phrases in speech recognitioncited by 1
  9. 2025Enhancing the Robustness of Contextual ASR to Varying Biasing Information Volumes Through Purified Semantic Correlation Joint Modelingcited by 1
  10. 2025PAC: Pronunciation-Aware Contextualized Large Language Model-based Automatic Speech Recognitioncited by 1
  11. 2026Dolphin-CN-Dialect: Where Chinese Dialects Mattercited by 1
  12. 2025H-PRM: A Pluggable Hotword Pre-Retrieval Module for Various Speech Recognition Systemscited by 0
6 more
  1. 2025VALLR-Pin: Uncertainty-Factorized Visual Speech Recognition for Mandarin with Pinyin Guidancecited by 0
  2. 2026From Oracle to Noisy Context: Mitigating Contextual Exposure Bias in Speech-LLMscited by 0
  3. 2026Development and multi-center evaluation of domain-adapted speech recognition for human-AI teaming in real-world gastrointestinal endoscopycited by 0
  4. 2026HearInContext: A Benchmark for Implicit Context in Speech Recognitioncited by 0
  5. 2026Can LLM Agents Automate Reinforcement Learning for Text-to-Speech?cited by 0
  6. 2026LLM-HB: Language-Aware LLM-Guided Hotword Biasing for Code-Switching ASRcited by 0