Audio Adversarial Examples for Robust Hybrid CTC/Attention Speech Recognition.
Ludwig KürzingerEdgar Ricardo Chavez RosasLujun LiTobias WatzelGerhard RigollPublished in: SPECOM (2020)
Keyphrases
- speech recognition
- speaker identification
- speech processing
- noisy environments
- speech recognition technology
- hidden markov models
- language model
- speech synthesis
- speech signal
- audio visual speech recognition
- pattern recognition
- speech understanding
- automatic speech recognition
- speech recognition systems
- speech recognizer
- multimedia
- cepstral coefficients
- keyword spotting
- acoustic features
- signal processing
- speaker independent
- audio signals
- speaker dependent
- speech retrieval
- speaker recognition
- probabilistic model
- multi stream
- speaker adaptation
- speech recognition errors
- broadcast news