A common problem for automatic speech recognition systems is how to recognize words that they did not see during training. Currently there is no established method of evaluating different techniques for tackling this problem. We propose using the CommonVoice dataset to create test sets for multiple languages which have a high out-of-vocabulary (OOV) ratio relative to a training set and release a new tool for calculating relevant performance metrics. We then evaluate, within the context of a hybrid ASR system, how much better subword models are at recognizing OOVs, and how much benefit one can get from incorporating OOV-word information into an existing system by modifying WFSTs. Additionally, we propose a new method for modifying a subword-based language model so as to better recognize OOV-words. We showcase very large improvements in OOV-word recognition and make both the data and code available.

翻译：自动语音识别系统的一个共同问题是,如何识别培训期间看不到的词句。目前没有评估解决这一问题的不同技术的既定方法。我们提议使用通用语音数据集为多种语言建立测试组,这些语言的外词汇比比高于一套培训,并发布新的工具,用于计算相关的性能指标。然后,在混合ASR系统的范围内,我们评估在承认OOOVs方面有多少更好的子词模型,以及通过修改WFST将OOOV词信息纳入现有系统能给谁带来多大好处。此外,我们提出了修改子字语言模型的新方法,以更好地识别OOOVS字。我们在OV-word字识别方面展示了很大的改进,并提供了数据和代码。