Lightweight and irreversible speech pseudonymization based on data-driven optimization of cascaded voice modification modules.
Hiroto KaiShinnosuke TakamichiSayaka ShiotaHitoshi KiyaPublished in: Comput. Speech Lang. (2022)
Keyphrases
- lightweight
- data driven
- text to speech
- speech recognition
- optimization problems
- speech synthesis
- voice activity detection
- emotion recognition
- speech signal
- fundamental frequency
- speech quality
- speech recognition errors
- automatic speech recognition
- authentication protocol
- optimization algorithm
- communication infrastructure
- speech sounds
- model driven
- rfid tags
- broadcast news
- wireless sensor networks