Martin, I can certainly address two of those comments. In terms of generating the tarball, it was not easy at all. I did it by writing a Python tool to sync the audio files from Voxforge, then configure and run sphinxtrain against them.
See: https://launchpad.net/voxforge-sphinx-train/ Is it possible to add a pointer to this program instead of automated rules? It takes >12 hours of computation time on a 8 core 16GB computer to do it also.. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1141942 Title: [needs-packaging] New package sphinx-voxforge-en To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+bug/1141942/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
