Speech Dispatcher 0.8.3 detailed changelog.

[Luke Yelavich]

Hey Folks.
For those who are interested, here is a detailed changelog for 0.8.3.

Felix Janda (2):
      Provide fallback for systems without TEMP_FAILURE_RETRY
      Make some includes consistent with POSIX

Jeremy Whiting (14):
      Add GET LANGUAGE support.
      Add GET RATE support.
      Add GET PITCH support.
      Add GET VOLUME support.
      ignore: Update .gitignore to ignore more generated files.
      style: Adjust all whitespace with ident.
      docs: Add documentation for new get apis.
      Parse: Simplify parse_get.
      python: Add methods for new get_foo methods to python client library.
      Added getter for voice_type. Added get voice_type to clibrary2 test.
      Add NEWLINE definition as "\r\n" and cleanup msg.h indentation.
      Fix Get methods by passing the SSIP name.
      Expand SPD_SET_COMMAND_INT|STR, SPD_GET_COMMAND_INT|STR
      Don't try to dereference voices or output_modules if they are null.

Luke Yelavich (15):
      Increment lib API version numbers due to newly added symbols
      include - Make sure to ship safe_io.h in the tarball
      Config - Drop GNOME speech client config file
      API - Fix up license for speechd types header
      API - Clean up voice/voice_type inconsistancies
      API - Provide version macros for major/minor/micro versions
      build - Bump libsndfile to a mandetory dependency
      Festival - Only terminate the speak thread if the thread has been created
      server - Use the GQueue data structure to remove hard-coded upper limits 
on voice lists
      espeak - Add config option to show variants in voice list
      Remove reference to GNOME Speech from the INSTALL file
      Forgot to update the NEWS file for 0.8.2
      python - Remove merge markers accidentally missed during a cherry-pick 
operation
      Update NEWS file for 0.8.3
      Update ANNOUNCE for 0.8.3

Luke