Join or Log in
Given an audio waveform, researchers can now produce a virtually identical version that makes speech-recognition software transcribe something else entirely. What's new: Berkeley researchers showed that they can take a waveform and add a layer of noise that fools DeepSpeech, a state-of-the-art speech-to-text AI, every time.
A new trick can spoof a speech recognition AI every time 
Added a month ago by Francis Tseng
Show info
A new trick can spoof a speech recognition AI every time 
Info
Given an audio waveform, researchers can now produce a virtually identical version that makes speech-recognition software transcribe something else entirely. What's new: Berkeley researchers showed that they can take a waveform and add a layer of noise that fools DeepSpeech, a state-of-the-art speech-to-text AI, every time.
1 Connection