We present a first large-scale public synthetic dataset for contextual
spellchecking customization of automatic speech recognition (ASR) with focus on
diverse rare and out-of-vocabulary (OOV) phrases, such as proper names or
terms. The proposed approach allows creating millions of realistic examples of
corrupted ASR hypotheses and simulate non-trivial biasing lists for the
customization task. Furthermore, we propose injecting two types of ``hard
negatives" to the simulated biasing lists in training examples and describe our
procedures to automatically mine them. We report experiments with training an
open-source customization model on the proposed dataset and show that the
injection of hard negative biasing phrases decreases WER and the number of
false alarms.

提出一个首个大规模的公开合成数据集，用于上下文拼写检查自定义自动语音识别（ASR）中的多样罕见和词汇外（OOV）短语，如专有名词或术语。通过注入两种类型的 “困难负面” 短语到模拟的偏见列表中，我们的方法可以创建数百万个真实的损坏 ASR 假设示例，并模拟定制任务的非平凡偏见列表，同时描述了自动挖掘它们的过程。通过在提出的数据集上训练开源定制模型进行实验，我们显示出注入困难负面偏见短语可以降低 WER 和误报数目。