## All
-   On supported platforms, the first time you select text and click the
    toolbar button, the extension reads the selected text aloud using
    default settings instead of showing a dialogue. This makes it
    quicker and easier for new users to get started.
-   Bug fixes and performance improvements.
-   On supported platforms, enable export of `x-sampa` or `ipa`
    International Phonetic data using an XML format. The [World Wide Web
    Pronunciation Lexicon
    Specification](https://www.w3.org/TR/pronunciation-lexicon/) (PLS)
    is a standard way to describe pronunciation. For example, accessible
    ePub books can use PLS files to help reading systems pronounce
    uncommon words, place names and personal names correctly.
    -   The PLS uses `alias` to refer to a sequence of words. For
        example, writers can use “W3C” to mean “World Wide Web
        Consortium”.
    -   The W3C Pronunication Lexicon Specification uses `phoneme` to
        refer to an alternate spelling that uses a phonetic code.
        -   [IPA](https://en.wikipedia.org/wiki/International_Phonetic_Alphabet)
            uses `/ˈkænjən/` for `canyon`. The strings can include
            special reserved IPA characters.
        -   [X-SAMPA](https://en.wikipedia.org/wiki/X-SAMPA) uses
            `["k{J@n]` for `canyon`. The strings use standard ASCII
            code.
    -   Specific language models might only support phonemes that are
        widely used in the language. As a result, they might not be able
        to pronounce some words correctly.
    -   Specific language models might require a specific phonetic
        code - like IPA or X-SAMPA.
    -   Each exported XML file includes only one regional pronunciation.
        Editors and publishers can optionally edit the file to specify
        pronunciation of specific words for other regions using a code
        editor.
## MacOS
The MacOS system allowed access to the phoneme editor even when phoneme
substitution was not available using the default AppleScript scripting
language. Now the extension disables right-click accessto editing
phonemes from the tools button of the *About* dialog if `python3` is not
enabled.
-   MacOS only supports phonemic substitution if `python3` is enabled in
    the main dialog. Some MacOS users might need to agree to the current
    “Xcode and Apple SDKs license” to use some features of this
    extension that use python code. You can view and agree or reject the
    Apple license agreement by entering the following in a terminal
    window: `sudo xcodebuild -license`.
-   Editing and using phonemic substitution requires you to enable
    `/usr/bin/python3` in the external program field of the main dialog
    of the extension dialog instead of the default `/usr/bin/osascript`.
## Windows
-   The file dialog for the saving a phoneme file did not suggest a file
    name. It now suggests a file name.
-   The extension can use StarBasic to generate SAPI text to speech if
    the Windows system `wscript.exe` program is not available.
## Linux
Make it easier to add and select
[piper-tts](https://github.com/rhasspy/piper#voices) voice models. Linux
Piper replaces Larynx, because Larynx contributors can no longer update
Larynx code.
-   The main menu includes some suggested voices for different
    languages.
-   When piper-tts is available, the main menu includes a menu item to
    update piper-tts metadata to include new languages and voices.