TY - BOOK ID - 8434655 TI - Self-learning speaker identification : a system for enhanced speech recognition AU - Herbig, Tobias. AU - Gerl, Franz. AU - Minker, Wolfgang. PY - 2011 SN - 3642198988 3642198996 PB - Berlin : Springer, DB - UniCat KW - Automatic identification. KW - Automatic speaker recognition. KW - Psycholinguistics. KW - Speaker adaptation. KW - Speech processing systems. KW - Electrical & Computer Engineering KW - Engineering & Applied Sciences KW - Telecommunications KW - Applied Physics KW - Electrical Engineering KW - Automatic speech recognition. KW - Mechanical speech recognizer KW - Speech recognition, Automatic KW - Engineering. KW - User interfaces (Computer systems). KW - Biometrics (Biology). KW - Electrical engineering. KW - Signal, Image and Speech Processing. KW - Biometrics. KW - Communications Engineering, Networks. KW - User Interfaces and Human Computer Interaction. KW - Electric engineering KW - Engineering KW - Biological statistics KW - Biology KW - Biometrics (Biology) KW - Biostatistics KW - Biomathematics KW - Statistics KW - Interfaces, User (Computer systems) KW - Human-machine systems KW - Human-computer interaction KW - Construction KW - Industrial arts KW - Technology KW - Statistical methods KW - Pattern recognition systems KW - Perceptrons KW - Speech, Intelligibility of KW - Speech perception KW - Speech processing systems KW - Telecommunication. KW - Computer science. KW - Informatics KW - Science KW - Electric communication KW - Mass communication KW - Telecom KW - Telecommunication industry KW - Communication KW - Information theory KW - Telecommuting KW - Signal processing. KW - Image processing. KW - Computational linguistics KW - Electronic systems KW - Modulation theory KW - Oral communication KW - Speech KW - Telecommunication KW - Singing voice synthesizers KW - Pictorial data processing KW - Picture processing KW - Processing, Image KW - Imaging systems KW - Optical data processing KW - Processing, Signal KW - Information measurement KW - Signal theory (Telecommunication) UR - https://www.unicat.be/uniCat?func=search&query=sysid:8434655 AB - Current speech recognition systems suffer from variation of voice characteristics between speakers as they are usually based on speaker independent speech models. In order to resolve this issue, adaptation methods have been developed in many state-of-the-art systems. However, information acquired over time is still lost whenever another speaker intermittently uses the recognition system. This work therefore develops an integrated approach for speech and speaker recognition in order to improve the self-learning opportunities of the system. A speaker adaptation scheme is introduced. It is suited for fast short-term and detailed long-term adaptation. These adaptation profiles are then used for an efficient speaker recognition system. The speaker identification enables the speaker adaptation to track different speakers which results in an optimal long-term adaptation. ER -