MycroftAI
/
mycroft-core
mirror of https://github.com/MycroftAI/mycroft-core.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
4,899 Commits
68 Branches
134 Tags
290 MiB
Commit Graph

212 Commits (refactor/python-detect)

Author SHA1 Message Date
Daniel McKnight eb97bb339e
Add `supported_languages` parameter to STT and TTS base classes (#3059) 
* Add `supported_languages` parameter to STT and TTS base classes

* Refactor `supported_languages` to `available_languages` and updated docstring to be more precise
 2022-02-28 15:36:48 +09:30
Kris Gesling 1ae6900b26
Remove ResponsiveVoice TTS module from core (#3049) 
This module in its current format does not work due to API changes.
There is also a community built plugin that is now described in our
documentation.
 2022-02-23 15:40:53 +09:30
Genei180 ef56d71462
Added Possibility for ESpeak Config (#3020) 
Added Possibility for ESpeak Config
 2021-11-09 15:07:33 +09:30
Kris Gesling 8a68c88bcf Add warning for anyone loading GoogleTTS 2021-08-27 09:49:40 +09:30
Åke Forslund 0ae02b5939 Fix TTS using the returned path 2021-07-02 08:34:03 +02:00
Kris Gesling 6099da18b3
Merge pull request #2934 from MycroftAI/bugfix/tts-audio-file-path 
Fix #2929: use returned TTS audio file path
 2021-06-30 07:03:01 +09:30
Kris Gesling 0394568704 Use returned TTS audio file path 
This reverts an unintentional breaking change.
A TTS engine may return a different file path than was requested.
This again uses the returned path but adds a deprecation warning
that this behaviour will no longer be supported in an upcoming release.

Fixes #2929
 2021-06-29 21:53:53 +09:30
Kris Gesling 21eabceb29 Do not create a 'None' cache directory 
The persistent_cache_dir was assigned None if no preloaded_cache existed.
This was then created as a directory when it's clearly not needed.
 2021-06-25 14:26:14 +09:30
Åke Forslund 5d3393b935 Make TTSCahce safer 
Add __contains__ method to TTSCache, The cache contains a SHA if the SHA is
known and all expected files exists on disk.

This is handles unexpected system events in a more consistent manner and
will still be fast for the case where a new sentence needs to be
synthetisized.
 2021-06-10 09:33:06 +02:00
Kris Gesling 1cc25da144
Merge pull request #2898 from MycroftAI/bugfix/docstring-cleanup 
Cleanup of documentation
 2021-06-01 12:16:44 +09:30
Kris Gesling 891784a454
Merge pull request #2888 from forslund/refactor/mimic_tts-dont-load-config-on-import 
Refactor mimic_tts to not load config when importing
 2021-05-28 15:24:13 +09:30
Kris Gesling e75a05557b docstring cleanup 2021-05-11 15:10:32 +09:30
Kris Gesling 0fb90edd28 Rename to get_temp_path and add docstring 2021-04-30 16:49:59 +09:30
dzekem christa 7765d11224 ran autopep8 2021-04-30 14:48:15 +09:30
dzekem christa 3f745c52ad created temp path function 2021-04-30 14:47:49 +09:30
dzekem christa 621b4ab650 spaces added and uniformity in code 2021-04-30 14:46:36 +09:30
dzekem christa 9252158829 rectified pep8 issues 2021-04-30 14:46:36 +09:30
dzekem christa 32f666edd4 Issue-2727 - fixing hard coded /tmp 2021-04-30 14:46:36 +09:30
dzekem christa a174c3c822 replaced hard coded /tmp 2021-04-30 14:46:36 +09:30
Åke Forslund d884da5b56 Be more defensive when loading mimic config 
Default config will generally always have a mimic block but in
combination with newer Mycroft variants a more defensive approach is
probably good
 2021-04-28 22:31:30 +02:00
Åke Forslund 5d68309523 Refactor mimic_tts to not load config when importing 
Hitting the entire configuration fetching routine with call to the
backend is not polite to do when just importing the file. This moves the
config lookup out of the global scope and into special functions for
finding the mimic binary and looking up the data path for the subscriber
voices.
 2021-04-28 22:31:10 +02:00
Åke Forslund 344999d003 Remove option to save phonemes in string format 
String format can't be loaded so it shouldn't be written either
 2021-04-05 08:28:38 +02:00
Åke Forslund bea3bac2bf Make mimic get_tts() return phonemes pre-parsed 
This makes the phonemes json de/encodable like mimic2
 2021-04-05 08:28:38 +02:00
Åke Forslund 0b857e92f9 Sync tts cache dict of cached sentences on curate 
This will make sure the dict of caches gets updated when curate_cache
removes files.
 2021-03-21 13:33:08 +01:00
Kris Gesling 7d3cb2d53e
Merge pull request #2869 from forslund/refactor/tts-cache-curation 
Move TTS cache curation to TextToSpeechCache
 2021-03-19 11:41:54 +09:30
Åke Forslund 9b4544dc7d Move cache curation to TextToSpeechCache 2021-03-18 23:02:26 +01:00
Åke Forslund 46a1de9b12 Remove unused import 2021-03-18 22:45:19 +01:00
Chris Veilleux 9c864de866
Convert phoneme to a list before being passed on. (#2865) 
* The phoneme needs to be converted to a list before being passed on.
 2021-03-18 15:23:20 +09:30
Chris Veilleux ea013b26af Fixed failing unit test. 2021-03-10 21:16:34 -06:00
Chris Veilleux 62d277d612 Refactored to fix error where persistent cache directory was used when the temporary cache directory should have been. 2021-03-10 21:16:11 -06:00
Chris Veilleux 469b1d3458 Changed ordering of adding a file to cache to make more sense. 2021-03-10 15:55:21 -06:00
Chris Veilleux 86febca6a3 Fixed a bug passing None to an instance of Path 2021-03-10 15:35:38 -06:00
Chris Veilleux 5996fdbf08 Fixed a minor linting issue 2021-03-10 13:46:48 -06:00
Chris Veilleux 48e691a66f add deprecation warnings to the logic the new cache logic replaces. 2021-03-10 13:39:30 -06:00
Chris Veilleux 2f34df6603 enhance to use new cache logic 2021-03-10 13:39:01 -06:00
Chris Veilleux 5e90940896 new code to consolidate TTS cache handling into one module. 2021-03-10 13:38:17 -06:00
Kris Gesling fa5576adf4
Merge pull request #2831 from MycroftAI/bugfix/google-voice 
Upgrade gTTS to fix Google voice
 2021-02-10 11:16:43 +09:30
Kris Gesling c8494932ac Remove deprecated Google TTS lang codes 2021-02-09 21:27:59 +09:30
Åke Forslund 9173e22460 Send correct conf to mimic when used as fallback 
- Mimic is now sent the correct config when used as a fallback
- Tests updated to ensure this
 2021-02-08 22:09:55 +01:00
Kris Gesling 5a19fc0716 Add Mozilla TTS connection validation 
Replication of PR #2719
Original code from JarbasAl
 2021-02-08 20:17:59 +09:30
Åke Forslund 082a2d481c Mark that audio has completed in dummy tts 2021-01-22 14:40:00 +01:00
Chris Veilleux 042a23879c Don't build the cache directory if it already exists. 2021-01-14 18:00:58 -06:00
Chris Veilleux a52d45ffff Re-added argument removed in previous commit to preserve backwards compatibility. 2021-01-14 14:29:44 -06:00
Chris Veilleux 91b1a9a15c Fixed an error that occurs when the pre-loaded cache directory already exists but the cache_text.txt file does not. 2021-01-13 21:47:09 -06:00
Kris Gesling 6f8bae6ba1
Merge pull request #2736 from forslund/docs/plugin-apis 
Add Plugin base classes APIs to readthedocs
 2020-12-03 20:33:59 +09:30
Kris Gesling b9c974bc1b
Merge pull request #2718 from forslund/bugfix/mimic-missing-exceptions 
Handle Mimic missing properly
 2020-11-16 14:40:14 +09:30
Åke Forslund 5e633421c6 Improve docstrings for TTS 2020-11-04 08:03:08 +01:00
Kris Gesling 8a1b989935
Merge pull request #2594 from forslund/feature/plugin-system 
Plugin system for stt, tts, audioservices and wake word engines
 2020-10-19 15:41:35 +09:30
Åke Forslund 112a983024 Fix pylint issues 2020-10-10 09:20:21 +02:00
Åke Forslund a2dc3482ef Fix logs when Mimic is not found 
If Mimic was not found at all the LOG message would raise an exception,
this fixes that issue and cleans up Exception chain
 2020-10-10 09:04:54 +02:00