The Sphinx4 speech recognition system is the latest addition to Carnegie Mellon University's repository of the Sphinx speech recognition systems. It has been jointly designed by Carnegie Mellon University, Sun Microsystems laboratories, Mitsubishi Electric Research Labs, and Hewlett-Packard's Cambridge Research Lab.
Sphinx 4 is different from the earlier CMUSphinx systems in terms of modularity, flexibility and algorithmic aspects. It uses newer search strategies, is universal in its acceptance of various kinds of grammars and language models, types of acoustic models and feature streams.
Algorithmic innovations included in the system design enable it to incorporate multiple information sources in a more elegant manner as compared to the other systems in the Sphinx family.
Sphinx 4 is developed entirely in the JavaTM programming language and is thus highly portable. Sphinx 4 also enables and uses multithreading and permits highly flexible user interfacing.
Work on the Sphinx 4 Trainer has begun.
The diagram below is the diagram initially proposed for Sphinx-4, but it has been refactored. The links point to their current location, but the blocks may have changed. Compare with the diagram on Figure 1 of the Sphinx-4 Overview page for a current diagram.
Notes from our face to face meetings:
The S4 team is made up of a number of individuals. The incomplete list is here. Please add yourself!