nvda

9cbe82e0 - Ensure the language code from a speech sequence is normalized before checking if the synth supports it. (#18541)

Ensure the language code from a speech sequence is normalized before checking if the synth supports it. (#18541) Summary of the issue: Recently, NVDA gained the ability to report the name of a language, when speaking text in that language and the synthesizer did not support it. However, in particular situations such as reading mathML, NvDA would report a language was not supported when it was. E.g. reading MathML in English with eSpeak, NvDA would report "English (not supported)". The reason for this seems to be that a non-normalized language code E.g. en-gb was being compared against normalized language codes, such as en_GB, and never matching. Description of user facing changes: NVDA will no longer incorrectly report a particular language is not supported. Description of developer facing changes: Description of development approach: In speech.languageHandling.getLangToReport: normalize the language.