H-PRM: A Pluggable Hotword Pre-Retrieval Module for Various Speech Recognition Systems

2025/08/22 by Dai, Huangyu, Mao, Lingtao, Chen, Ben +5
#Artificial Intelligence (cs.AI) #Audio and Speech Processing (eess.AS) #Computation and Language (cs.CL) #FOS: Computer and information sciences #FOS: Electrical engineering #Sound (cs.SD) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2508.18295

Abstract

Hotword customization is crucial in ASR to enhance the accuracy of domain-specific terms. It has been primarily driven by the advancements in traditional models and Audio large language models (LLMs). However, existing models often struggle with large-scale hotwords, as the recognition rate drops dramatically with the number of hotwords increasing. In this paper, we introduce a novel hotword customization system that utilizes a hotword pre-retrieval module (H-PRM) to identify the most relevant hotword candidate by measuring the acoustic similarity between the hotwords and the speech segment. This plug-and-play solution can be easily integrated into traditional models such as SeACo-Paraformer, significantly enhancing hotwords post-recall rate (PRR). Additionally, we incorporate H-PRM into Audio LLMs through a prompt-based approach, enabling seamless customization of hotwords. Extensive testing validates that H-PRM can outperform existing methods, showing a new direction for hotword customization in ASR.

Citations

• CTC-Assisted LLM-Based Contextual ASR
• Qwen2-Audio Technical Report
• A Multitask Training Approach to Enhance Whisper with Contextual Biasing and Open-Vocabulary Keyword Spotting
• SeACo-Paraformer: A Non-Autoregressive ASR System with Flexible and Effective Hotword Customization Ability
• Two Stage Contextual Word Filtering for Context bias in Unified Streaming and Non-streaming Transducer
• Robust Speech Recognition via Large-Scale Weak Supervision
• Paraformer: Fast and Accurate Parallel Transformer for Non-autoregressive End-to-End Speech Recognition
• Improving End-to-End Contextual Speech Recognition with Fine-Grained Contextual Knowledge Selection
• Improving non-autoregressive end-to-end speech recognition with pre-trained acoustic and language models
• Fast Contextual Adaptation with Neural Associative Memory for On-Device Personalized Speech Recognition
• An Improved Single Step Non-autoregressive Transformer for Automatic Speech Recognition
• Personalization Strategies for End-to-End Speech Recognition Systems
• CIF-based Collaborative Decoding for End-to-end Contextual Speech Recognition
• CASS-NAT: CTC Alignment-based Single Step Non-autoregressive Transformer for Speech Recognition
• Spike-Triggered Non-Autoregressive Transformer for End-to-End Speech Recognition
• Conformer: Convolution-augmented Transformer for Speech Recognition
• Conformer: Convolution-augmented Transformer for Speech Recognition
• Deep context: end-to-end contextual speech recognition
• Attention Is All You Need
• Listen, Attend and Spell
• Attention-Based Models for Speech Recognition
• Speech Recognition with Deep Recurrent Neural Networks
• Sequence Transduction with Recurrent Neural Networks

Related

• Contextual Biasing for LLM-Based ASR with Hotword Retrieval and Reinforcement Learning
• SeACo-Paraformer: A Non-Autoregressive ASR System with Flexible and Effective Hotword Customization Ability
• CTC-Assisted LLM-Based Contextual ASR
• RAG-Boost: Retrieval-Augmented Generation Enhanced LLM-based Speech Recognition
• LA-RAG:Enhancing LLM-based ASR Accuracy with Retrieval-Augmented Generation