| Document Date: 2010-04-12 23:27:20
Open Document File Size: 444,69 KBShare Result on Facebook
City Denver / SPEECH GMM / London / Taipei / / Company Wall Street Journal / SLM / / Country Taiwan / / Currency USD / / / Facility Building Synthetic Voices / Carnegie Mellon University / Karlsruhe University / Language Technologies Institute / / IndustryTerm subsequent processing / feature processing / automatic speaker identification systems / de-identification technologies / speaker identification systems / bank / automated voicedriven applications / null-grammar network / energy / / MarketIndex set 95 / / MusicAlbum Speech / Vol.3 / / Organization American Medical Informatics Association / Language Technologies Institute / Carnegie Mellon University Pittsburgh / Vol.14 / Karlsruhe University / / Person Phonetic Model / Arthur R. Toth / Voice Convergin / Identification / Alan W Black / Qin Jin / Large / Tanja Schultz / / Position single target speaker / Language-dependent Speaker / original speaker / source speaker / same speaker / Gaussian Mixture Speaker / target speaker / Speaker / enrolment speaker / ARCTIC speaker / modified speaker / Text-independent Speaker / Decision LSPM Recognizer decoding Scoring Logic Language-dependent Speaker / source and target speaker / / ProvinceOrState Vermont / / PublishedMedium the Wall Street Journal / / Technology de-identification technologies / technology of voice de-identification / / URL http /
SocialTag |
Speaker De-identification via Voice Transformation Qin Jin #1, Arthur R. Toth #2, Tanja Schultz #3, Alan W Black #4 # Language Technologies Institute, Carnegie Mellon University Pittsburgh, PA 15213, USA