You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, to get the language-specific SAMPA form of each phoneme, a working (and espeak-friendly) installation of mbrola is required.
This is problematic for several reasons:
it requires an additional system-wide package install on linux platform (even though mbrola is available on most distributions)
it requires the installation of a corresponding mbrola voice, which is either unpractical and/or quite heavy. Moreover, the entirety of the voice's speech data isn't actually used by espeak for the phonemization.
Currently, to get the language-specific SAMPA form of each phoneme, a working (and espeak-friendly) installation of mbrola is required.
This is problematic for several reasons:
The foldings are all here: https://github.com/espeak-ng/espeak-ng/tree/master/phsource/mbrola
The text was updated successfully, but these errors were encountered: