AM-MobileNet1D: A Portable Model for Speaker Recognition

03/31/2020 ∙ by João Antônio Chagas Nunes, et al. ∙ 0

Speaker Recognition and Speaker Identification are challenging tasks with essential applications such as automation, authentication, and security. Deep learning approaches like SincNet and AM-SincNet presented great results on these tasks. The promising performance took these models to real-world applications that becoming fundamentally end-user driven and mostly mobile. The mobile computation requires applications with reduced storage size, non-processing and memory intensive and efficient energy-consuming. The deep learning approaches, in contrast, usually are energy expensive, demanding storage, processing power, and memory. To address this demand, we propose a portable model called Additive Margin MobileNet1D (AM-MobileNet1D) to Speaker Identification on mobile devices. We evaluated the proposed approach on TIMIT and MIT datasets obtaining equivalent or better performances concerning the baseline methods. Additionally, the proposed model takes only 11.6 megabytes on disk storage against 91.2 from SincNet and AM-SincNet architectures, making the model seven times faster, with eight times fewer parameters.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 1

page 2

page 3

page 4

Code Repositories

AM-MobileNet1D

The Additive Margin MobileNet1D is a new light weight deep learning model for Speaker Recognition which is based on the MobileNetV2 architecture and the Additive Margin Softmax (AM-Softmax) loss function.)


view repo
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.