BriefGPT.xyz
Oct, 2020
重新思考ASR中的评估:我们的模型足够健壮吗?
Rethinking Evaluation in ASR: Are Our Models Robust Enough?
HTML
PDF
Tatiana Likhomanenko, Qiantong Xu, Vineel Pratap, Paden Tomasello, Jacob Kahn...
TL;DR
研究探讨了在自动语音识别中单一基准测试数据的数字推动是否具有价值,发现噪声增强有助于提高模型的泛化性能,并且使用大量的基准测试数据可以良好地代表真实世界中的性能表现,最终得出在广泛使用的数据集上训练单一声学模型可达到竞争性的研究和现实世界基准测试性能。
Abstract
Is pushing numbers on a single
benchmark
valuable in automatic speech recognition? Research results in
acoustic modeling
are typically evaluated based on performance on a single dataset. While the research commun
→