Sound masking for achieving speech privacy with parametric acoustic array speaker.
Takahiro TamesueTetsuro SaekiPublished in: SCIS&ISIS (2014)
Keyphrases
- acoustic features
- speaker verification
- automatic speech recognition
- speech signal
- sound source
- speech recognition
- audio visual
- prosodic features
- speech sounds
- environmental sounds
- mel frequency cepstral coefficients
- information loss
- automatic speech recognition systems
- acoustic models
- visual features
- speaker recognition
- music information retrieval
- speech recognition systems
- speaker independent
- privacy preserving
- speech synthesis
- noisy environments
- vocal tract
- hidden markov models
- audio stream
- speaker identification
- speech recognizers
- private information
- privacy preserving data mining
- sensitive information
- human visual system
- personal information
- speaker dependent
- statistical databases
- emotion recognition
- differential privacy
- speaker diarization
- noisy speech
- visual speech
- privacy protection
- pattern recognition
- privacy issues
- audio signal
- parametric models
- audio features
- security issues
- speech segments
- spontaneous speech
- text to speech
- privacy policies