kttsmgr dosen't work

please I need help with my opensuse 11.3 64 bit, I have installed “festival” and “kttsmgr” but when i try to configure kttsd i cannot choose any speaker because the list appear empty.
Festival by shell works.
Can anyone help me to get voice by my desktop?
Thanks,
Tonino

It isn’t just you. I have been struggling with this for months. Somehow it has fallen through the cracks. 11.2 worked, but 11.3 is borked. There are missing paths and it is over my head to fix. I have gotten zero help from anyone so far. Maybe you will have more luck.

KTTS is not ready yet. You can write your own application, which seems to be the only alternative right now. I am using the CMU Pocketsphinx speech recognition engine which requires some adaptation or the replacement of the language model. I have almost 0 accuracy on recogition right now. On top of that your microphone and sound card will affect performance, as well as the settings in alsamixer.

I have found that the sound card must be set in alsamixer so it can be adjusted. By default it is set to default which gives almost no settings. When I selected card 0, all sorts of adjustment possibilities showed up. Additionally, you will need Audacity as a recorder to make recordings if necessary, or just to test the system.

Alsamixergui is useless as is krecord, in my current experience.

If you decide to try Pocketsphinx, python-devel needs to be installed first if you wish to write an application with Python. Otherwise, you would have to learn the Java that was used to create the engine. I have not gotten there yet, but my intent is to use python since it is easier to read the code. I suspect there would be much reading while developing even a simple system.

The Sphinx wiki is very good on Sourceforge and they have a forum. Do not expect them to teach you programming, however. They may help you with programming if you get stuck at a certain point, but only if you demonstrate that you have put some effort into your project, and that you have a basic understanding of the programming language.

As far as tts, basically, Festival will speak whatever is in a text file that you have created. To make it speak,

festival --tts [path to the file]

I have found Kaffeine renders the voice better than Amarok. I set my default player for .wav files to Kaffeine. I never checked to see what format Festival renders its voice, but I changed all .wav files to play by default with Kaffeine and it works. I did this to test an arctic voice that I downloaded. I could not deal with ked_diphone.

Jovie works

You are right 11.3 I could not get it to work.
11.4, I now have my system speaking the time to me every 30 minutes like I had in 11.2. , and clip board etc.
Hint= Ktts was replaced with Jovie. I have a big list of talkers in jovie.
It took a little bit to figure out what software to install to get it to work. It is all in the repo’s.

Thank you for the tip. I had hoped ktssmgr would allow me to send commands to the system by my voice. Is Jovie able to do that?

I have no idea if it can. I don’t think so. My guess is you are looking for some Linux Voice Recognition software.

I did a Google search and found quite a bet on it.

Good luck.

Yes there is quite a bit on it but there is no plug and play. I must be built. The tts is there in Festival rpm. Sphinx must be installed manually and is not easy to get it anywhere near 100% recognition. At least it installs and functions on openSUSE, fortunately.

Dragon has a version which permits all kinds of custom commands for the os, but it costs $600 and will not run on Linux without Wine. I would do neither of these, so I guess I must learn Python. Check back with me in a few months. Hopefully, I will have a program that will send commands to the system.