The OS X TTS engine (and likely others) allows outputting its speech directly
to the sound system. This avoids the extra step of creating a temporary file to
play for TTS preview. Currently implemented as TTS capability reported.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@29672 a1c6a512-1295-4272-9138-f99709370657
Note that some voices don't support speed adjustment. These voices will ignore
the selected value. The current implementation doesn't mark those voices in any
way.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@25125 a1c6a512-1295-4272-9138-f99709370657
This adds basic support for using the OS X' TTS system. The current
implementation does only support selecting the voice itself, no further
settings (like speed pitch / speed) adjustments are implemented. As OS X' TTS
system wants the strings to get spoken in 8 bit encoding problems with locale
combinations are possible. For this better error handling in the rbutil TTS is
needed. The voice test button in the configuration dialog reacts pretty slow
due to the way its speaking is done. This can get changed but also requires
adjustments in the rbutil TTS system.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24979 a1c6a512-1295-4272-9138-f99709370657
