User-visible changes to speechd-el:

* Changes in speechd-el 2.11

** Index marking support

Set speechd-speak-use-index-marks variable to `t' to enable index
marking.  Note that it may have performance or other issues.  How well
index marking actually works is also dependent on the speech
synthesizer used.

** Improvement of reading wrapped lines

If truncate-lines is nil, lines are read only to their visual ends
(thanks to Nicolas Graner).

** Internal cleanup

The code has been updated to match current Emacs practices and all
compilation warnings have been resolved.


* Changes in speechd-el 2.10

** New user option speechd-speak-emacs-language.

It sets the language for messages read in English by default.

** Bug fixes.


* Changes in speechd-el 2.9

** Bug fixes.


* Changes in speechd-el 2.8

** XDG_RUNTIME_DIR environment variable is honored now.

Thanks to Alain Kalker for this change.


* Changes in speechd-el 2.7

** Minor bug fixes.

Thanks to Alain Kalker for help.


* Changes in speechd-el 2.6

** Minor bug fixes.

Thanks to Kevin Ryde for identifying the bugs.

** Packaging support.


* Changes in speechd-el 2.5

This is a maintenance release, primarily fixing problems with newer Emacs
versions.


* Changes in speechd-el 2.4

** Support for new Speech Dispatcher autospawn mechanism.

speechd-el uses Speech Dispatcher autospawn feature by default, see
`speechd-autospawn' user option.

** Support for new Unix socket communication with Speech Dispatcher.

Unix socket communication is used by default, see `speechd-connection-method'
user option.

** Isearch reading adjustments.

** New user option speechd-speak-echo.

It allows you to set what speechd-el speaks when you type characters.


* Changes in speechd-el 2.3

** Buffer names in speechd-speak-connections are regexps now.

** Braille key handling improvements.

** Automatic restarts of broken SSIP and BrlAPI connections.


* Changes in speechd-el 2.2

** Support for new BrlAPI protocol introduced in BRLTTY 3.8.

speechd-el now works with BRLTTY 3.8 final version and BRLTTY 3.9.
Additionally BRLTTY error reporting cleanup was made.

** New BRLTTY key handler for general character input.

New macro `speechd-brltty-command-key-function' is provided to assign actions
to Braille display general input keys, such as letters.

** New command speechd-set-synthesizer-voice.

It allows setting speech synthesizer voice directly by its name.  Note this
only works with Speech Dispatcher 0.6.3 or higher.

** Moved to GPL3.


* Changes in speechd-el 2.1

** Invisible text is no longer read.

** Support for BRLTTY 3.8.

BRLTTY uses 64-bit key codes.  They are represented as two-number lists in
speechd-braille-key-functions.

** WINDOWPATH BRLTTY environment variable is recognized.

The WINDOWSPATH environment variable should be named WINDOWPATH now.
This speechd-el version uses whichever of those two is defined.


* Changes in speechd-el 2.0

** Output mechanism modularized.

speechd-el is output device independent now, it no longer depends on speech
output and SSIP.  It is now possible (and sufficiently simple) to add new
output modules for alternative output devices, such as Braille displays, etc.

** BRLTTY input/output added.

speechd-el now supports Braille output to BRLTTY with basically the same
functionality as in the speech output.  It is possible to use speech output,
Braille output, or both.  You can use sophisticated speechd-el features such as
message priorities and text property handling instead of screen reading.
Additionally it is possible to bind functions to the Braille display keys.

** Braille display emulator added.

There is another output module available: Braille display emulator in an Emacs
buffer.  Useful only for development purposes.

** Message manager added.

Message manager for managing messages and priorities was added.  It works in a
way similar to the Speech Dispatcher message mechanism.


* Changes in speechd-el 1.0

** The top customization group is called speechd-el now.


* Changes in speechd-el 0.5

** Messages from built-in functions can be read now.

** New user option speechd-speak-message-time-interval.

It allows you to control speaking of repeated messages.

** Mode line can be read in CVS versions of Emacs.

See the command speechd-speak-read-mode-line bound to the `C-e RET' key.

** New commands providing basic information about current environment.

Type `C-e C-i C-h' to see the list of those commands.

** Some state changes can be automatically reported now.

See the variable speechd-speak-state-changes.

** The SPEECHD_HOST environment variable is honored now.

** Changes in communication with Speech Dispatcher.

Former socket based asynchronous communication with Speech Dispatcher was
unreliable and could cause errors in some situations.  This has been fixed for
CVS Emacs versions.  In Emacs 21.3, two alternative workarounds are available:
either one-way communication with Speech Dispatcher (the default) or using the
spdsend utility (if the speechd-spdsend variable is set).

** speechd-connection-parameters removed.

This obsolete variable is no longer available, use speechd-connection-voices
instead.


* Changes in speechd-el 0.4

** Whitespace signalization added.

It allows you, among others, to distinguish between empty lines and lines
containing only whitespace.

** Repetition of last speakings possible in character reading prompts.

For instance, it is possible to repeat Ispell choices in ispell-word now.

** Language selection based on currently selected input method is possible.

See the new variable speechd-speak-input-method-languages.

** Asynchronous action handling reworked.

The former mechanism of speaking in asynchronous actions, such as process
outputs or timers, was broken.  It should be fixed now.

** Responsiveness improved.

speechd-el is now much more responsive when speaking many small pieces of text.


* Changes in speechd-el 0.3

** Connection parameters are handled in a different way.

Connection parameters are configured as voices in the speechd-voices variable
now.  Each specific voice can then be assigned to a connection or face, see the
variables speechd-connection-voices and speechd-face-voices.  The variable
speechd-connection-parameters is obsoleted and shouldn't be used anymore.

** Bug reporting changes.

Bug reproduction finishing key was changed from `C-e C-z' to `C-e C-f'.
New command speechd-bug-reproduce allowing to reproduce a bug any time.
Log extraction speed up.

** Volume setting added.

You can use the `C-e d V' command with Speech Dispatcher 0.3 or higher to
adjust speech volume.

** The *Completion* buffer is spoken now.

** You can shut up speechd-el in debugger.

See the new variable speechd-speak-in-debugger.


* Changes in speechd-el 0.2

** read-event prompts are spoken now.

** Spelling support added.

You can now use speechd-speak-spell-mode to enable spelling in a buffer or the
`C-e C-l' keys to cause the next command to spell the text it reads.

** Choosing widget values improved.

The variable widget-menu-minibuffer-flag is now automatically set to t when
speechd-speak-mode is enabled, thus the widget value is chosen using usual
minibuffer completions.

** Bug reproduction finishing key was changed from `C-e .' to `C-e C-z'.


Local variables:
mode: outline
end: