## All
- Dispose of speech status file locks after a crash or an unexpected
system reboot.
- Improve readability and update code.
## Linux
- Update default [piper-tts] lexicons.
- If you install the [pied] speech manager using the latest tar.gz
file from the [GitHub website] in a local directory instead of using
a [flatpak] or [snap][GitHub website] package manager, the extension
can now find the location of the pied speech synthesis resource
files.
- When using a local docker image of [mimic3-server] text to speech,
ignore voices with improperly installed .onnx speech data.
- Use [piper-tts] by default when available instead of [festival].
- Show the (SPD_READ_TEXT_PY) option whenever it would produce speech,
even if your [spd-say] text to speech engine does not support the
current language.
## Windows
- Update default [piper-tts] lexicons.
- Fix a delay when displaying the lexicon editor dialog.
[piper-tts]: https://github.com/rhasspy/piper/blob/master/README.md
[pied]: https://pied.mikeasoft.com/
[GitHub website]: https://snapcraft.io/pied
[flatpak]: https://flatpak.org/
[mimic3-server]: https://mycroft-ai.gitbook.io/docs/mycroft-technologies/mimic-tts/mimic-3
[festival]: https://manpages.debian.org/testing/festival/festival.1.en.html
[spd-say]: https://manpages.debian.org/testing/speech-dispatcher/spd-say.1.en.html