Sonogram visible speech has been programmed at the German Research Center for Artifical Intelligence (Deutsches Forschungszentrum für Kuenstliche Intelligenz DFKI, www.dfki.de), and is a 3D tool to analyze speech and sound signals with STFT and various other algorithms. The software originates from basic research and the attempt to build a speech recognizer with neuronal networks. The new version requires Java at least in version 16. If you have any questions or problems running or compiling the Sonogram, don't hesitate to contact us. The program was written with emacs.
Contact: Christoph Lauer, [email protected]
Note: You have to be Java at least in version 16 installed on your Computer in order to use Sonogram. See Oracle Java or AdoptOpenJDK. Apple Silicon (ARM) users should use the native Azul Zulu JDK or JRE.
If you have trouble with the Sonogram settings you can start Sonogram with CAPS-LOCK enabled.
a reset screen is shown:
You can build Sonogram by you own. First you have to install a java SDK at least in version 16. Make sure Java is in your PATH variable.
cd source
make all
Note: If you determine memory problems (for example in the 3D generator) give the java VM more heap memory with the xmx flag in the corresponding startup scrips and the makefile.