Oct 15 2020
Oct 1 2020
Sep 17 2020
Need to update to latest version of mishkal
Sep 8 2020
Changed to use
logging.handlers.SysLogHandler(address = '/dev/log')
Added a setting to wikispeech-server/default.conf:
#logger defaults to stderr, uncomment here if you want something else
There is now an endpoint /default_voices that returns json similar to this:
Output json now contains timing in milliseconds instead of seconds
audio_url is no longer returned.
Tmp dir is cleared after each call.
Sep 3 2020
Sep 2 2020
Jul 23 2020
Jul 20 2020
Yes, wrong format was returned. Now fixed in git.
Jul 10 2020
Json item "voice", with items adapter, config_file, engine, language, name has been added.
Jun 26 2020
Apr 2 2020
Partly done - a call to the wikispeech server, eg. http://localhost:10000/?lang=sv&input=sju%20sj%C3%B6sjuka%20sj%C3%B6m%C3%A4n. now also contains audio_data, a base64 encoded string. The audio_url still there as well, will remove that later. For testing you can also try adding the parameter output_type=html, eg. http://localhost:10000/?lang=sv&output_type=html&input=sju%20sj%C3%B6sjuka%20sj%C3%B6m%C3%A4n.
This will return very simple html with an audio element containing the base64 string as source.
Mar 19 2020
I can't reproduce this - here's what I get when starting MaryTTS without a running mishkal server:
Feb 6 2020
Nov 28 2019
Nov 26 2019
The reason for the "spike" is still unknown. But the low volume after is caused by the "Amplitude Normaliser" in Marytts.
Nov 25 2019
No longer relevant. Close.
But if we're still using marytts we still need to consider the speed issue later.
Already fixed. Close.
Already fixed. Close.
Seems fine - open again if something like this happens again!
This is part of the (complicated) question of how to deal with ambiguity in pronunciation and part-of-speech tags. If there is a "real" pos-tagger at work, or if pos-tags are set by the user, these should be used to select pronunciation. Otherwise the lexicon might just as well set it. Changed (test) to set part-of-speech in the same way as pronunciation has been set before: Use the first entry in the lexicon, unless another one is explicitly set as preferred.