Anonymizing Speech with Generative Adversarial Networks to Preserve Speaker Privacy.
Sarina MeyerPascal TilliPavel DenisovFlorian LuxJulia KochNgoc Thang VuPublished in: SLT (2022)
Keyphrases
- speech recognition
- social network data
- speaker recognition
- private information
- audio visual
- privacy protection
- speaker verification
- automatic speech recognition
- data anonymization
- speaker identification
- social networks
- privacy preserving
- information loss
- vocal tract
- speaker diarization
- speech signal
- speaker dependent
- automatic speech recognition systems
- noisy environments
- text to speech
- personal information
- acoustic features
- gaussian mixture model
- social network analysis
- prosodic features
- generative model
- multi agent
- privacy preserving data mining
- network structure
- speech recognizer
- privacy preservation
- feature vectors
- speech sounds
- automatic transcription