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

TL;DR

SeACo-Paraformer is a novel non-autoregressive ASR system that enhances hotword customization, combining accuracy, efficiency, and explicit customization, validated on large-scale industrial data.

Contribution

It introduces a new NAR-based ASR model with improved hotword customization and training stability, outperforming existing baselines and providing open-source tools.

Findings

Outperforms strong baselines in hotword customization accuracy

Efficient filtering method improves hotword handling

Validated on 50,000 hours of industrial data

Abstract

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.