EUROSPEECH 2001 Scandinavia
The purpose of this paper is to introduce a user-friendly computer assisted learning (CAL) workbench in order to support traditional teaching in the Speaker Recognition area. The workbench (an interactive on-line laboratory) is based on Java and Java-enabled Web browser. The first prototype demonstrator developed at the Swiss Federal Institute of Technology Lausanne (EPFL) for speaker identification training consists of four modules: Dynamic Time Warping (DTW), hidden Markov Modelling (HMM), Vector Quantization (VQ) and Gaussian Mixture Modelling (GMM). These four modules aim at presenting, visualizing and investigating in a uniform way basic concepts of speaker recognition in a single user-friendly environment and allow for easy and highly illustrative learning through experiments with real speech data. They can be used for conventional classroom experiments, in the students' laboratory or can provide self-study means for distance learning applications or for further free exploration.
Bibliographic reference. Drygajlo, Andrzej / Garcia Molina, Gary (2001): "Javaspeakerrecognition - interactive workbench for visualizing speaker recognition concepts on the WWW", In EUROSPEECH-2001, 2787-2790.