File name problem during model creation when input wav is not PCM16

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Use not PCM16 wav file as input for model creation 
2. vid -s test -g test_not_PCM16.wav


What is the expected output? What do you see instead?

[FeatureSet] read : compute data test java.lang.IllegalArgumentException: 
Unsupported conversion: PCM_SIGNED 16000.0 Hz, 16 bit, mono, 2 bytes/frame, 
big-endian from PCM_SIGNED 44100.0 Hz, 16 bit, stereo, 4 bytes/frame, 
little-endian
    at javax.sound.sampled.AudioSystem.getAudioInputStream(AudioSystem.java:955)
    at fr.lium.spkDiarization.libFeature.FeatureFactory.getAudio(FeatureFactory.java:161)
    at fr.lium.spkDiarization.libFeature.FeatureFactory.MakeFeature(FeatureFactory.java:191)
    at fr.lium.spkDiarization.libFeature.FeatureFactory.MakeMFCCFeature(FeatureFactory.java:234)
    at fr.lium.spkDiarization.libFeature.FeatureSet.read(FeatureSet.java:740)
    at fr.lium.spkDiarization.libFeature.FeatureSet.setCurrentShow(FeatureSet.java:1091)
    at fr.lium.spkDiarization.system.Diarization.ester2Version(Diarization.java:310)
    at fr.lium.spkDiarization.system.Diarization.main(Diarization.java:540)

What version of the product are you using? On what operating system?

Latest trunk, Mac OX X 10.7.4

Please provide any additional information below.

The problem is that fm.py:file2wav converts source input (test.wav) to 
test_.wav and then
test._wav is not used for later diarization. The solution could be in renaming 
source file before conversation
to e.g. test_orig.wav and then save conversation result with name of initial 
source file name e.g. test.wav

Original issue reported on code.google.com by aritomo....@gmail.com on 24 Oct 2012 at 1:24

[GoogleCodeExporter]

This issue was closed by revision r236.

Original comment by maurome...@gmail.com on 30 Oct 2012 at 1:08

• Changed state: Fixed