MARF is a general cross-platform framework with a collection of algorithms for audio (voice, speech, and sound) and natural language text analysis and recognition, along with sample applications (identification, NLP, etc.) of its use. MARF can run distributed over the network (CORBA, Java RMI, and Java XML-RPC Web Services) and may act as a library in applications or be used as a source for learning and extension.
|Tags||education Testing multimedia Sound/Audio Analysis Conversion Speech Scientific/Engineering Artificial Intelligence Software Development Libraries Java Libraries|
|Operating Systems||OS Independent|
Release Notes: A bug has been fixed in Sample.setSampleSize() where the if() condition would always evaluate to false when checking whether the resulting sample is shorter or longer than the original. A bug in custom Sample serialization where the order of writing an int and the array data was reversed has been fixed. The comments have been updated a bit more for the updates. Some related unit tests have been added/fixed in order for the failures not to reoccur. The marf mini-app's purpose has been clarified.
Release Notes: This is a bugfix release that corrects NullPointerException in various places and option handling.
Release Notes: A code review was done. Training data (compiled training sets) are included. The usage instructions in the README file were updated. All options were tested. Assertions were enabled.
Release Notes: This release included a code review, updated scripts, training.sh and testing.sh files, an updated README, a version check, and enabled assertions.
Release Notes: This release features an option for FE aggregation, training sets, enabled assertions, more samples, an -endp option for Endpoint, a testing.bat script, a 'retrain' Windows shortcut that runs the whole process of training and testing, --batch-ident and --single-train options, refactored code, and use of OptionProcessor.