| Age | Commit message (Collapse) | Author |
|---|
|
* Return pitch -= 6 to females, this brings the pitch closer
to zero
* Add array of indices which ensures, that we use as many voices
as we can.
|
|
* Add checks if ttsMan != null before trying to use it
* Simplify startSpeech
* Move haut to startSpeech and pass the character index to
the startSpeech instead.
|
|
Remove debuging outputs
Remove commented iconv implementation of text conversion to UTF-8
|
|
The waitSpeech should use the old code, when just sound is playing
(the TTS isn't speaking).
|
|
|
|
|
|
|
|
|
|
This was broken in previous commit when adding support for versions
that do not have this file.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
the German version
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
display
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Using a LoopingAudioStream gets rid of the pause between each loop
(and simplifies the code).
|
|
This was due to the _speakerHandle being reused between its
initialisation in the SoundManager constructor and the destructor
causing it to have a wrong value when trying (and failing) to stop
the PCSpeaker channel in the mixer before deleting the stream.
|
|
|
|
the song
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
alternate french versions
|
|
|
|
|
|
|
