The following forum message was posted by mokhov at http://sourceforge.net/projects/marf/forums/forum/213052/topic/4878330:
Hi safatron, sorry for the delay in reply. Thanks for the good comments! Now to your specific comments: > it is necessary to write down a sound from a microphone in wav a file, > and then it to load for the further processing. In the traditional MARF and its SpeakerIdentApp, yes. > It does not turn out in any way. What do you mean? Do you mean you want to put the input stream directly for processing without creating an intermediate file first? It is possible and some (external) applications have done that, but you'd still be storing some information somewhere, like training sets, processed utterances, etc. The real-time recording then would be a part of the application itself. Inside MARF, we began work long ago to add a SampleRecorder class [1], which would get the data not from the file, along with its Test application [2], but it was not fully finished with other things taking more priorities. If you feel like, you are welcome to contribute the missing bits ;-) [1] http://marf.cvs.sourceforge.net/viewvc/marf/marf/src/marf/Storage/SampleReco rder.java [2] http://marf.cvs.sourceforge.net/viewvc/marf/apps/TestSampleRecorder/ BWT, what kind of scientific works do you do? Any links? There's been several publications with MARF and derived works and applications. Thanks, -s ------------------------------------------------------------------------------ Learn Windows Azure Live! Tuesday, Dec 13, 2011 Microsoft is holding a special Learn Windows Azure training event for developers. It will provide a great way to learn Windows Azure and what it provides. You can attend the event by watching it streamed LIVE online. Learn more at http://p.sf.net/sfu/ms-windowsazure _______________________________________________ marf-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/marf-devel
