Anonymizing Speech with Generative Adversarial Networks to Preserve Speaker Privacy.
Sarina MeyerPascal TilliPavel DenisovFlorian LuxJulia KochNgoc Thang VuPublished in: CoRR (2022)
Keyphrases
- speech recognition
- social network data
- speaker recognition
- automatic speech recognition
- privacy protection
- private information
- audio visual
- social networks
- data anonymization
- speaker identification
- speech signal
- speaker verification
- data publishing
- vocal tract
- privacy preserving
- social network analysis
- speaker diarization
- information loss
- gaussian mixture model
- community structure
- privacy preservation
- speaker dependent
- multi modal
- automatic transcription
- multi agent
- automatic speech recognition systems
- speech synthesis
- prosodic features
- acoustic features
- broadcast news
- sensitive information
- generative model
- data privacy
- location privacy
- privacy concerns
- visual speech
- speaker independent
- community detection
- personal information
- network structure
- visual information
- synthesized speech