Open Source Toolkit Documentation:

Audio Echo Example Using SoX

License / Documentation home / Help and feedback

Description

This example is almost identical to the audio echo example, except that the SoX wrapper and its associated GUI are used instead of the audio echo server. Because SoX is threaded by necessity, it does not need to run in threaded mode as the audio echo server must.

Starting up

Unix

% <OSTK_HOME>/examples/cats_sox/script/cats_sox.csh

Windows

This example does not yet run on Windows.

Running

Start the servers in order. When the Hub connects to the audio GUI server, two windows will appear. See the audio GUI server documentation for more details about how to use this GUI. Note especially that the audio GUI may require JDK 1.4 in order to run reliably. When the Hub connects to the SoX GUI server, the SoX GUI window will appear.

Once all the servers are up and the audio GUI is started, press the "Off Hook" button on the keypad. The initial configuration is full-duplex, streaming, with barge-in and VAD enabled. You should be able to control each of these parameters in the controller GUI.

Initially, the SoX GUI is configured as a pass-through, and provides functionality identical to the audio echo server. Because it's inherently not as efficient, it introduces more of a lag in streaming mode than the audio echo server. Once you've established connectivity, you can try a few things:

Select Add... -> Output filename and provide an output filename, then press Send Settings, and then Resend in the audio GUI. The audio will be saved to the specified file.
Now select Add... -> Input filename and provide the same filename, then press Send Settings. SoX will send the audio in the file to CATS. You can delete the output filename setting line if you like, but it doesn't matter if there's still an output filename setting; the program file ignores all the output settings if an input filename is present.
Now delete the output and input filenames, select Add... -> Volume and provide a value of 0.1, press Send Settings, and press Resend in the audio GUI. The output should be faint.
Now delete the volume line, and select Add... -> Effect, select reverb from the list of effects, and type "1.0 200 100 50" in the input entry, and press Send Settings, and then Resend in the audio GUI. You should hear reverb added to the voice.