Comparing human and automatic speech recognition in simple and complex acoustic scenes.
Computer Speech & Language（2018）
Abstract
•Automatic speech recognition and human listeners are compared in single-channel and spatial scenes.•In single-channel scenes, ASR is on a par with normal-hearing listeners.•In spatial scenes, there is a substantial human-machine gap of 12.3 dB.•5.3 dB of this gap can be attributed to poor localization and missing speaker-related features.
MoreTranslated text
Key words
Human-machine comparison,Speech recognition threshold,Deep neural networks,Speech intelligibility prediction,Spatial scenes
AI Read Science
Must-Reading Tree
Example
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined