Evaluating deep learned voice compression for use in video games
作者：
Highlights：
• Identification of the key unique characteristics in the game audio compression.
• Outline the dataset requirements to adequately test video game quality audio.
• Propose and perform objective performance measures on video game audio data.
• Provide initial results on off-the-shelf versions of the algorithms.
• Evaluate any gaps that need to be addressed.
摘要
•Identification of the key unique characteristics in the game audio compression.•Outline the dataset requirements to adequately test video game quality audio.•Propose and perform objective performance measures on video game audio data.•Provide initial results on off-the-shelf versions of the algorithms.•Evaluate any gaps that need to be addressed.
论文关键词：Voice compression,Deep learning,Video games,Objective voice quality measurement,Performance
论文评审过程：Received 4 December 2019, Revised 13 April 2021, Accepted 8 May 2021, Available online 16 May 2021, Version of Record 19 May 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.115180
