I updated to the svn version in the trunk but still no luck. I really don't know...
I'm using the same acoustic model and experience the same problem. Here are the output...
I'm using the same acoustic model and experience the same problem. Here are the output...
I'm using the same acoustic model and experience the same problem. Here are the output...
Somehow there was another pocketsphinx version installed on the system. After reinstall...
Hello, Im switching from Sphinx4 to Pocketsphinx since MLLR is not possible in Sphinx4....
Hello, Im switching from Sphinx4 to Pocketsphinx since MLLR is not possible in Sphinx4....
Hello, Im switching from Sphinx4 to Pocketsphinx since MLLR is not possible in Sphinx4....
Hello, I found some demo code like that one: https://www.assembla.com/code/sonido/subversion/nodes/3/sphinx4/src/apps/edu/cmu/sphinx/demo/jsapi/tags/FeatureValueDemo.java...
Okey that makes sense. The audio file is pretty unaccurate to test the behaviour...
Is it possible to get the most likely result although the accuracy is very low?
After this change i don't get any hypothesis in the second case.
sphinx4-5prealpha
So how do i force to only consider the whole rule as a result?
Hello, I'm just getting started with sphinx and adopted the transcriber demo to use...
Hello, I'm just getting started with sphinx and adopted the transcriber demo to use...