Could not find or load main class org.openhab.voice.voicerss.tool.CreateTTSCache

Add-ons Text-to-Speech
,
1

I am using voicerss to send some messages over squeezebox (OH2 is running on VM ubuntu 16.04, latest updates, latest snapshots, java 1.8.0_151) using the say command (just picks a message randomly out of the a selection of 20 or so).
In general, this works well, when the mp3 files are available, but the ‘say’ command times out often (though not always) when no cached mp3 is available.

So I thought pre-populating the cache may take care of this, and there are instructions in the docs: https://docs.openhab.org/addons/voices/voicerss/readme.html

Usage: java org.openhab.voice.voicerss.tool.CreateTTSCache <args>
Arguments: --api-key <key> <cache-dir> <locale> { <text> | @inputfile }
  key       the VoiceRSS API Key, e.g. "123456789"
  cache-dir is directory where the files will be stored, e.g. "voicerss-cache"
  locale    the language locale, has to be valid, e.g. "en-us", "de-de"
  text      the text to create audio file for, e.g. "Hello World"
  inputfile a name of a file, where all lines will be translatet to text, e.g. "@message.txt"

Sample: java org.openhab.voice.voicerss.tool.CreateTTSCache --api-key 1234567890 cache en-US @messages.txt

When I typed the sample command above in a console, I got this error:
Error: Could not find or load main class org.openhab.voice.voicerss.tool.CreateTTSCache

OH2 runs well, so I figured that it is not path problem, but obviously I am missing something.
Any and all thoughts will be appreciated

2

I know it’s a bit old, but I also have the same problem. Did you find a way to fix it?

Thanks!!

3

I was never really sure what resolved my issues; maybe updating helped, maybe the internet gremlins are more willing to allow for that…
but the bottom line, i have not seen my original issue in quite some time (even after having deleted the cache) so I never pursued that any further. My next step would have been to pre-compile on startup (or on request through say a proxy item), that is I envisaged to run through all the strings at start-up (probably with the volume all the way down…:slight_smile: )