``SPEAKER''

The speaker-tag now works with german and english. Speakers of another language may easily be added.
As default language (between the end of one and the beginning of another speaker-tag) the language, selected by the user before analyzing the sable file, is used.
Possible values

• for english speakers are:
  • male1
  • male2
  • male3
  • female1
  • male4
  • male5
• for german speakers are:
  • de1
  • de2
  • de2_linginto
  • de3_linginto
  • (julia_pizzas)
  • (julia_drinks)
  • (mark_numbers)

I'm not sure, if they all really work!
It is now also possible to choose only the gender of the speaker by using: <SPEAKER GENDER="female" ...> or <SPEAKER GENDER="male" ...>. Here, we also differ only between german and english, but other languages may easily be added by inserting another cond-branch equivalent to the german one in the GENDER-branches at the opening of the speaker-tag.
Just a few things to be done to add speakers of another language (see language-tag):

1. ``init-function'':
   1. another branch has to be added to the cond-statement equivalent to the ``german-branch''
   2. in ``sable_init_globals'', another branch has to be added to the cond-statement equivalent to the ``german-branch''
2. ``restore_ttw_functions'':
   1. another branch has to be added to the cond-statement equivalent to the ``german-branch''
3. ``set_ttw_functions'':
   1. another branch has to be added to the cond-statement equivalent to the ``german-branch''

How the speaker-tag works:

• opening of the speaker-tag:
  1. xxml_synth
  2. current speaker is pushed on the sable_speaker_stack
  3. set back token_to_words-function of current language (at the moment only distinguished between german and non-german) using ``restore_ttw_functions''
  4. selection of the new speaker
  5. set token_to_words on sable_token_to_words because this is overwritten in the setting of english voices
  6. set german_token_to_words on sable_token_to_words (after saving german_token_to_words in sable_previous_token_to_words) if language of chosen voice is german
     using ``set_ttw_functions''
     set english_token_to_words on sable_token_to_words (after saving english_token_to_words in sable_previous_token_to_words) if language of chosen voice is not german
     using ``set_ttw_functions''
  7. sable_setup_voice_params
• closing of the speaker-tag:
  1. xxml_synth
  2. set back token_to_words-function of current language (at the moment only distinguished between german and non-german) using ``restore_ttw_functions''
  3. pop last speaker from stack
  4. evaluate speaker popped from the stack
  5. set token_to_words on sable_token_to_words because this is overwritten in the setting of english voices
  6. set german_token_to_words on sable_token_to_words (after saving german_token_to_words in sable_previous_token_to_words) if current language is german
     using ``set_ttw_functions''
     set english_token_to_words on sable_token_to_words (after saving english_token_to_words in sable_previous_token_to_words) if current language is not german
     using ``set_ttw_functions''