用于端到端语音识别模型文本自适应的快速准确因子分解神经变换器
Fast and accurate factorized neural transducer for text adaption of end-to-end speech recognition models
摘要：
神经传感器由于其自然的流式传输能力，成为目前最流行的语音识别端到端模型。然而，它是具有挑战性的，以适应它与纯文本数据。因子化神经传感器（FNT）模型的提出，以减轻这个问题。与标准神经传感器模型相比，FNT对纯文本适应数据的适应能力的提高是以降低准确性为代价的。我们提出了几种方法来提高FNT模型的性能。它们是：在训练过程中增加CTC准则，在自适应过程中增加KL发散损失，使用预训练的语言模型来播种词汇预测器，以及通过用n-gram语言模型内插词汇预测器的有效自适应方法。这些方法的组合结果在一个相对的字错误率从标准的FNT模型减少了9.48%。此外，使用词汇预测器的n-gram插值极大地提高了自适应速度，具有令人满意的自适应性能。
英文摘要：
Neural transducer is now the most popular end-to-end model for speech recognition, due to its naturally streaming ability. However, it is challenging to adapt it with text-only data. Factorized neural transducer (FNT) model was proposed to mitigate this problem. The improved adaptation ability of FNT on text-only adaptation data came at the cost of lowered accuracy compared to the standard neural transducer model. We propose several methods to improve the performance of the FNT model. They are: adding CTC criterion during training, adding KL divergence loss during adaptation, using a pre-trained language model to seed the vocabulary predictor, and an efficient adaptation approach by interpolating the vocabulary predictor with the n-gram language model. A combination of these approaches results in a relative word-error-rate reduction of 9.48\% from the standard FNT model. Furthermore, n-gram interpolation with the vocabulary predictor improves the adaptation speed hugely with satisfactory adaptation performance.
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。
