You will need to install python development library (not only python binary package). If you use apt-get, the command will be like $sudo apt-get install python2.7-dev The package name could slightly different. You may want to confirm the exact package name with "apt-cache search | grep python |grep dev" We are switching from gnu make to cmake. This install process will be much easier in btk2.0.
hi, I found that when I configure, it report the following error. Can anyone help me out, Thanks a lot! configure: error: Could not link test program to Python. Maybe the main Python library has been installed in some non-standard library path. If so, pass it to configure, via the LDFLAGS environment variable. Example: ./configure LDFLAGS="-L/usr/non-standard-path/python/lib" ========================================= ERROR! You probably have to install the development version of the Python package...
Hi, BTK/ASR~~~ Thank you for your fast response. It's difficult for me to progress this project now. Good luck everybody. Bye...
The doc is missing some details (needs to be updated). The readDirectoinOfArrival() is not defined there. It just is supposed to read the incident angle between the array and direction of interest from a file. What you need is to set the incident angle in radians to "avgDOA". If your target sound is right in front of the linear array, the incident angle will be pi/2.
Hello, Good morning... I am also watching this project, and analyze the Delay-and-Sum Beamforing of User Manual. But I met 1 problem at this sentence , The error message is below. from btk.beamformer import * pBeamformer = SubbandGSCPtr( fftLen=M, halfBandShift=False) for chX in range(chanN): ... pBeamformer.setChannel(allAnalysisFBs[chX]) ... for chX in range(chanN): ... allSampleFeats[chX].read( fn='KINECT_M1001_U1050_2M.wav', chX=(chX + 1), chN=chanN )... 89600 89600 89600 89600 import numpy sampleRate...
You will need to generate it with speaker tracking module. A sample speaker tracking script is contained in Scripts_DS+SD+MVDR.tar.gz "/KSFTspeakerTrackingEKF.py" implements Kalman-filering-based speaker tracking.
btk home
btk home
btk home
Home
Hi I am also watching this project recently. Can I leave a contact form for quick and easy discussion? It’s too deserted here.
I find its the direction fo arrival which can be calculate bt shape tracking. Thanks a lot.
Hello: I see the example of delay-and-sum Beamforming in the manual. In the example, there is the file name direction.txt which shown in '>>avgDOA = readDirectoinOfArrival('direction.txt' )". I wonder how can I generate this direction.txt and what is the format of direction.txt?
thank you MaYong. I had the same problem and your solution worked!
Solved it. Had to change Makefile.am, Makefile.in and configure file and insert "dereverberation"...
After following the isntruction in the manual and installing, the import dereverberation...
After following the isntruction in the manual and installing, the import dereverberation...
@Tomo did you find any resolution for this?
UPDATE: in 1.: these "fade-out" effects are not caused by dereverberation, but directly...
Hi, I'm trying to employ Your dereverberation library to improve our speech recognition....
Hi, I'm trying to employ Your dereverberation library to improve our speech recognition....
Hi, I'm trying to employ Your dereverberation library to improve our speech recognition....
problem sovled: it maybe that because my compiler version is different. I sovled...
hi, I found that when I make, it report the following error. ../stream/pyStream.h:120:56:...
hi, I found that when I make, it report the following error. ../stream/pyStream.h:120:56:...
Fixed a include path problem in Makefile.am.
Added a source detection flag.
Replaced the old python scripts.
Replaced the old python scripts.
Fixed Makefiles.
Replaced the old python scripts.
Added a dereverberation component.
Removed unnecessary files.
Initial commit
Add initial directories