When the Differences in Frequency Domain are Compensated: Understanding and Defeating Modulated Replay Attacks on Automatic Speech Recognition.
Shu WangJiahao CaoXu HeKun SunQi LiPublished in: CoRR (2020)
Keyphrases
- frequency domain
- automatic speech recognition
- speech recognition
- spatial domain
- fourier transform
- power spectrum
- speech signal
- conversational speech
- denoising
- fourier analysis
- word error rate
- hidden markov models
- speech retrieval
- cross correlation
- broadcast news
- frequency analysis
- feature extraction
- spectrum analysis
- speech corpus
- noisy environments
- power spectra
- discrete fourier transform
- watermarking scheme
- face recognition
- computer vision