Hacker News new | past | comments | ask | show | jobs | submit login
CommanderSong: Practical Adversarial Voice Recognition (2018) (arxiv.org)
18 points by JoachimS 29 days ago | hide | past | web | favorite | 2 comments



Speech to text still feels hard to get. Every now and then I try to find a codebase to do this locally on a pi but am always disappointed.


The abstract seems to indicate it's a way to attack speech recognition systems (Alexa, Google Home) by embedding commands into music?




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: