arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.12258cs.CLcs.LG
收藏

序列到序列学习的语音识别错误的幻觉

Hallucination of speech recognition errors with sequence to sequence learning

The Ohio State University(俄亥俄州立大学)

Prashant Serai, Vishal Sunder, Eric Fosler-Lussier

摘要:

自动语音识别(ASR)是一个不完美的过程,当与纯书面文本或transmittance相比时,它会导致ASR输出文本中的某些不匹配。当纯文本数据被用于训练口语语言理解或ASR的系统时,减少所述失配并防止降级的经证实的策略是使ASR输出的内容产生幻觉,将被给予黄金转录。该领域的先前工作集中在语音级别的建模错误,同时使用词典将音素转换为单词,通常伴随着FST语言模型。我们提出了新的端到端模型,直接预测幻觉ASR单词序列输出,条件输入单词序列以及相应的音素序列。这改进了先前公布的结果,用于从域内ASR系统对未看见的数据的转录以及域外ASR系统对来自不相关任务的音频的转录中召回错误,同时另外探索当可获得来自测试ASR系统的有限表征数据时的中间场景。为了验证该方法的外在有效性,我们还使用我们的幻觉ASR错误来增强口语问题分类器的训练,发现当训练时特定于任务的音频很少或甚至为零时,它们能够对下游任务中的真实ASR错误具有鲁棒性。

英文摘要:

Automatic Speech Recognition (ASR) is an imperfect process that results in certain mismatches in ASR output text when compared to plain written text or transcriptions. When plain text data is to be used to train systems for spoken language understanding or ASR, a proven strategy to reduce said mismatch and prevent degradations, is to hallucinate what the ASR outputs would be given a gold transcription. Prior work in this domain has focused on modeling errors at the phonetic level, while using a lexicon to convert the phones to words, usually accompanied by an FST Language model. We present novel end-to-end models to directly predict hallucinated ASR word sequence outputs, conditioning on an input word sequence as well as a corresponding phoneme sequence. This improves prior published results for recall of errors from an in-domain ASR system's transcription of unseen data, as well as an out-of-domain ASR system's transcriptions of audio from an unrelated task, while additionally exploring an in-between scenario when limited characterization data from the test ASR system is obtainable. To verify the extrinsic validity of the method, we also use our hallucinated ASR errors to augment training for a spoken question classifier, finding that they enable robustness to real ASR errors in a downstream task, when scarce or even zero task-specific audio was available at train-time.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑