Using automatic speech recognition for attacking acoustic CAPTCHAs: the trade-off between usability and security.
Hendrik MeutznerViet-Hung NguyenThorsten HolzDorothea KolossaPublished in: ACSAC (2014)
Keyphrases
- automatic speech recognition
- trade off
- acoustic features
- speech sounds
- speech recognition
- speech recognizers
- information security
- speech segments
- speech signal
- acoustic models
- hidden markov models
- word error rate
- speech recognition systems
- speech retrieval
- spoken words
- recognition errors
- formant frequencies
- word recognition
- broadcast news
- noisy environments
- discriminative training
- spontaneous speech
- speech recognizer
- speech corpus
- sound source
- speaker independent
- conversational speech
- language model