Knowledge Commons of Institute of Automation,CAS
Hypersphere Embedding and Additive Margin for Query-by-example Keyword Spotting | |
Ma Haoxin1,2; Bai Ye1,2; Yi Jiangyan1; Tao Jianhua1,2,3 | |
2019-11 | |
会议名称 | APSIPA 2019 |
会议日期 | 2019-11 |
会议地点 | 中国兰州 |
摘要 | Query-by-example (QbE) keyword spotting is convenient for users to define their own keywords, so it is useful in device control. However, conventional regular softmax, which is commonly used for training QbE models, has two limitations. First, the learned features are not discriminative enough. Second, norm variations of the unnormalized features affect computing cosine similarities. To address these issues, this paper introduces normalization and additive margin into residual networks for QbE keyword spotting. Features and weights are normalized on a hypersphere of fixed radius. Additive margin further helps to reduce the intra-class variations and increase inter-class differences. Based on public datasets AISHELL-1 and HelloNPU, we design three different test sets, namely in-vocabulary, out-of-vocabulary, and cross-corpus, to evaluate our proposed method. Experiments show that our proposed method can learn more discriminative embedding features. For totally unseen situation, our proposed method achieves a relative false rejection rate reduction of 46.60% when the false alarm rate is 2% in cross-corpus evaluation, compared with regular softmax. |
收录类别 | EI |
语种 | 英语 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/48841 |
专题 | 多模态人工智能系统全国重点实验室_智能交互 |
作者单位 | 1.NLPR, Institute of Automation, Chinese Academy of Sciences, China 2.School of Artificial Intelligence, University of Chinese Academy of Sciences, China 3.CAS Center for Excellence in Brain Science and Intelligence Technology, China |
第一作者单位 | 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Ma Haoxin,Bai Ye,Yi Jiangyan,et al. Hypersphere Embedding and Additive Margin for Query-by-example Keyword Spotting[C],2019. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Hypersphere_Embeddin(3993KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论