Year: 2021
Type: conference-paper
Abstract: In this paper, we propose a deep convolutional neural network-based acoustic word embedding system for code-switching query by example spoken term detection. Different from previous configurations, we combine audio data in two languages for training instead of only using one single language. We trans-form the acoustic features of keyword templates and searching content segments obtained in a sliding manner to fixed-dimensional vectors and calculate the distances between them. An auxiliary variability-invariant loss is also applied to training data within the same word but different speakers. This strategy is used to prevent the extractor from encoding undesired speaker- or accent-related information into the acoustic word embeddings. Experimental results show that our proposed sys-tem produces promising searching results in the code-switching test scenario. With the employment of variability-invariant loss, the searching performance is further enhanced.
Institutions: Duke Kunshan University, Lenovo (China), Wuhan University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: G2099609658, G8933391830
OpenAlex ID: W3134903552
