Leveraging Domain Features for Detecting Adversarial Attacks Against Deep Speech Recognition in Noise.
Christian Heider NielsenZheng-Hua TanPublished in: CoRR (2022)
Keyphrases
- speech recognition
- noisy environments
- speech recognition systems
- language model
- hidden markov models
- automatic speech recognition
- speech signal
- cepstral coefficients
- speech synthesis
- speech recognition technology
- pattern recognition
- feature extraction
- isolated word
- information retrieval
- speech recognizer
- speech processing
- speech recognizers
- audio visual speech recognition
- speaker identification
- mel frequency cepstral coefficients
- low level
- speech retrieval
- signal to noise ratio
- feature vectors
- feature space
- face recognition
- image processing