- 12 source files
- 1 makefile
- 1 external component(executable)
Click here to select download mirror.
The Tk interface needs ActiveTCL to be installed at your end.
If you have wish or tsh at you end already, then you need to change the first line of the interface script to point it to the location of the wish or the tsh .
You can start by extracting all the files in a single directory. You may call this new directory as "vox".
Running the makefile will in turn create an executable "vox" and two directories "tmp" and "repositories" in the working directory.
The system is command line based. So the exact syntax of its usage can be obtained by typing the command
$./vox
in the current directory.
To create a repository you need a phonetically balanced training file which is basically a speech file in MS WAVE format recorded at 8000 samples per second with 16 bits per sample of a duration of 15 minutes. (The repository generation phase may take a long time to complete.)
To encode you need a message file which is basically a speech file in MS WAVE format recorded at 8000 samples per second with 16 bits per sample.
If for some reason the program aborts abruptly during the repository generation phase, then you need to install the Edinburgh Speech Tools Library's sig2fv tool which needs to be placed in the working directory before execution. You may follow this link to download the required executable.
Edinburgh Speech Tools Library
For more details refer to the FAQ and documentation.
Please respond with your invaluable suggestions to the project admins. For contact details, see this.
Thanking you for downloading.