File: history.html

package info (click to toggle)
transcriber 1.5.1.1-11
links: PTS, VCS
area: main
in suites: sid, trixie
size: 4,484 kB
sloc: tcl: 13,141; ansic: 1,895; sh: 1,193; makefile: 96; xml: 66
file content (1093 lines) | stat: -rw-r--r-- 38,756 bytes
parent folder | download | duplicates (2)
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html><head>
   <meta content="text/html; charset=iso-8859-1" http-equiv="Content-Type">
   <meta content="Mozilla/4.78 [en] (X11; U; Linux 2.4.7-10 i686) [Netscape]" name="GENERATOR">
   <meta content="C. Barras" name="Author"><title>Transcriber's History</title></head><body alink="#ff0000" vlink="#51188e" link="#0000ef" bgcolor="#ffffff" text="#000000">

<h2>
History of Transcriber versions</h2>

<hr width="100%"><b><font size=" 0">Version 1.5.1</font></b>

<p>Changes or bugs fixed:<br>
<ul>
<li> Upper extensions are now recognized as valid extensions for transcription and audio files. </li>
<li> A DTD different from trans13.dtd or trans14.dtd is not considered anymore as a former DTD. </li>
<li> All unicode encoding are now supported on Windows </li>
<li> Transcriber 1.5.0 refused to be launched if it found an external speaker database. That file was created when the speaker turns were modified. That bug has been corrected. </li>
<li> The default browser (used for viewing help files) could not be set. This bug has been corrected. </li>
</ul>
<p> </p>
<hr width="100%"><b><font size=" 0">Version 1.5.0</font></b>

<p>New or modified functions:<br>
</p>
<ul>
  <li>A new DTD (trans-14.dtd) replace the old one (trans-13.dtd). For
  compatibility, files that need an older DTD can be opened with
  Transcriber-1.5.0. For such files, save or exit without any
  changes will display a message that inform you that the original
  format of your file is older than the format of your Transcriber
  application. You can choose to export in new format (recommended) or
  keep the original one.</li>
  <li>improvement of the "save audio
  selection" function: the menu entry is now "Save audio segment(s)"
  with two submenus "Selected..." and "Automatic...". You can also
  reached this function with contextual menu by right clicking on the
  signal and look in "Audio file" entry. The first submenu do exactely
  the same as the old function (i.e. save an audio selection) and the
  second one is an automatic mode to save all the audio segemnts
  corresponding to section, turn and sync. A dialog box allow you to
  choose wich kind of segment do you want and where you want to save
  it in.</li> <li>the event tags can now be translated when displayed
  in Transcriber, but they are not translated in TRS files. </li>
  <li>the binary released 1.5.0 is now available on Linux,
  Windows</li> <li>the spell checker is now Aspell and works on Linux
  and Windows </li> <li>help files are now viewed with the default
  browser </li> <li>shape computation for large files is now available
  on Windows </li> <li>help files are available in German</li>
  <li>now, when you destroy a begin or end event it also destroy the
  symetric associated event.</li> <li>trs files are associated to
  Transcriber on Windows</li>
  
  <li>shortcuts F1 to F7 are available. F1 is for help, F2 to F7 are
  used to hide or not the different parts of the interface (text, NE,
  command button, first and second signal view and message
  display).</li>

  <li>creation of a new event "Named entities" for named entities (NE)
  annotation. The F3 button is a shortcut that make available an
  interface to easily annotate NE. Automatic annotion is possible and
  every colors or fonts are configurable. The list of entities can
  also be modified. More details in the reference manual.</li>

  <li>creation of a global speakers database which
  contains all the global speakers you have imported in it. Import is
  done by Edit->Speakers->Update global speakers database, then you
  have to select the file from wich you want to import global
  speakers. More details in the reference manual.</li>

  <li>transcription files may use a different format than the original
  one provided with the package, provided that they point to a DTD
  located in the same directory, that their root element is still
  &lt;Trans&gt; and that they are in a superset of the existing
  format.  Unknown XML tag will be only displayed in gray, but may not
  be edited.  Any XML file whith &lt;Trans&gt; root tag is now
  supposed a valid .trs file, regardless of its actual extension (it
  used to be limited to .trs and .xml).<br> </li> 
  
  <li>it is possible to define a default topic or speaker import file,
  loaded when opening a transcription or creating a new one, using
  'Edit/Speaker|Topic/Automatic import from selected file'<br> </li>

  <li>export a transcription to text format in a simple "Speaker i:
  transcription" format; control over the formatting may be done by
  setting some configuation values on the command line or in the
  configuration file - see convert/text.tcl for more details (NB: a
  beta version of this filter was already distributed in the
  1.4.6)</li> 
  
  <li>display of NIST RT'03S .mdtm format (speaker subset
  only).</li>

  <li>added "-version" command line option.<br> </li>

</ul>
<p>
</p>

<p>changes or bug fixes:<br>
</p>
<ul>
  <li>removing of lrec documentation (useless) </li>
  <li>removing  of html_library (useless) </li>
  <li>Migration of user preferences to Transcriber 1.5.0 (Event widget size is reset)</li>
	<li>popups Find/Replace, Find Speaker, Find Topic, Spell Checker, Control Panel, Informations are now in foreground</li>
	
  <li>a warning window pops up if more than one transcription line is deleted</li>
  <li>fixed bug : an error message appeared when doing File/Synchronized Audio File/Add Audio File. That made the function unvailable.</li>

  <li>fixed bug : on some platform when using the Num_lock key, the edition was impossible.</li>

  <li>fixed bug : when changing localisation (for example english to french) the contextual menu of the wave were not translate so that when you wanted to open an audio file, an error message appreared.</li>

  <li>a warning message is displayed when temporal inconsistencies are found in a .trs file.</li><li>Export
to .stm format now includes a commented header giving information about
the conversion process, and the file is written in the user-defined
encoding (rather than the default system encoding) - it may be possible
to override it on the command line e.g with 'trans -set encoding utf-8
-export stm *.trs'.<br>
  </li>
  <li>multiwav extension now has a menu entry in the main 'File/Synchronized audio files' menu.</li>

  <li>'Open segmentation file..' has been moved frome 'Options/Display' to&nbsp; 'File' menu.</li>
  <li>default string for new speaker or topics may be modified
(currently only by editing the default settings or the configuration
file, see etc/default.txt).<br>
  </li>
  <li>improved parsing of LDC hub4 files in sgml/utf formats for segmenation display.<br>
  </li>
  <li>bug corrected in the "-set" line option.</li>
  <li>bug corrected in saving a modified localization file.<br>
  </li>
  <li>command line option "-export" may be used instead of "-convertto".</li><li>Transcriber
is launched in batch mode (i.e. with tclsh rather than wish) by 'trans'
script when -export or -help command line options are found.<br>
  </li>

  <li>under Tcl/Tk 8.4, it should be possible to open multiple files in a single dialog for the multiwav and segmentation features.</li>
  <li>convert filters using the extended tcl API 'readSegmtSet' have been made backward compatible with previou versions (&lt;1.4.6).<br>
  </li>
</ul>
<p>
</p>
<hr width="100%"><b><font size=" 0">Version 1.4.6</font></b><p>Released on October 3, 2003. <br>

<p>New or modified functions:
</p>
<ul>
	    <li>multiwav extension, allowing several synchronized audio
	      files to be associated with a single transcription. New audio
	      files can be added using the contextual menu on the waveform
	      (right-click), or a list of audio files can be given on the
	      command line when launching the tool.
	      It is derived from an original proposal by David Gelbart and
	      Dan Ellis from ICSI, intended for meeting recordings management.</li><li>Czech localization of the menus added, kindly provided by Zdenek Konopasek.</li>
  <li>Management of several configuration files (menus "Options/Save
configuration as..." and "Options/Load configuration file..."); a
configuration file may be specified at startup with '-cfg filename'
command line option.</li>
  <li>Save audio selection in a separate .wav file (menu "File/Save audio selection as...").</li>
  <li>Explicit message when signal shape is not available and resolution display is thus limited.<br>
  </li>
  <li>Text editor may be hidden (checkbox menu "Options/Display/Text editor", '-notext' command line option).</li>
  <li>Segmentation layers for background, sections or turns which are
not in use (i.e. having a single segment for the whole document) may be
automatically hidden (checkbox menu "Options/Display/Smart segmentation display").</li>
  <li>Random colorization of the turn segmentation layer, so that a
color is automatically attributed to all occurences of the same speaker
in the document (checkbox menu "Options/Display/Colorize speaker segments"). </li>
  <li>Open a label file and display under the signal (menu
"Options/Display/Open segmentation file..." or command line option
"-lbl filename" -- option may be repeated). The label file information
may span several segmentation layers.<br>
  </li>
  <li>Improved support of NIST .ctm file (including RT'03 format
extensions and multi-level display as labels); .stm can be either imported or displayed as multi-level labels.<br>
  </li>
  <li>Added '-convertto format filename ...' command line option for
converting a set of file from the Transcriber format to the given
format in batch mode.<br>
  </li>
  <li>Various other command line options added: '-set option value' for
overriding configuration value, '-sig2' for display of second signal
view, '-socket' for activation of external scripting service, '-help'
for a list of options.<br>
  </li>
</ul>changes or bug fixes:
<ul>
<li>added compatibility update for Tcl8.4, due to changes in (internal) Tcl
	      and C API. Transcriber binary libs will probably be incompatible between
	      Tcl versions 8.3 and 8.4</li><li>added support for Mac OS X (Darwin/Aqua).</li>
  <li>extended conversion filter API with 'readSegmtSet', for management of multi-level label files.<br>
  </li>
  <li>discard unreadable default signal or transcription files from a configuration file.<br>
  </li>
  <li>localization of the interface is now in a separate file for each language, for a better extensibility.</li>
  <li>segmentation layer name in the signal contextual menu becomes more explicit ("Background", "Turns", etc.)<br>
  </li>
  <li>resolved a confusion involving GB-2312 encoding<br>
  </li>
</ul><hr width="100%"><b><font size=" 0">Version 1.4.5</font></b>
<p>Released on July 11, 2002.
<br>needs Tcl/Tk 8.0-8.3, Snack 1.7 or preferably 2.0.1, tcLex 1.2
</p><p>New or modified functions:
</p><ul>
<li>
added new playback mode stopping before (rather than after) next boundary;
added related menu entry</li>

<li>
added menu entry and ctrl-return binding for playing next segment</li>

<li>
added bindings alt-backspace/delete for deleting word before/after cursor
respectively.</li>

<li>
added ctrl-alt-t binding for modifying turn attributesImplementation.</li>

<li>
added support for floating boundaries.&nbsp;They are viewed in the text
editor as empty circles (standard boundaries being displayed as full bullets)
and it is possible to toggle the state of each boundary or of the whole
set of boundaries by right-clicking (or control-left-clicking) on the bullet
(except the first one). When a boundary is moved, the neighbouring floating
ones move while keeping a proportionality up to the next "hard" boundary.&nbsp;A
floating boundary becomes automatically fixed after being moved.&nbsp;The
floating state is marked in the .trs format with a space before the time
value (this behaviour is meant for backward compatibility of .trs files
with prior versions of Transcriber and should evolve in the future). Floating
boundaries are intended for later support of import and edition of non-temporized
manual transcriptions or dialogs.</li>

<li>
added import filters for various formats defined by NIST speech group (/www.nist.gov/speech/),
especially the output of the 'sclite' scoring tool (alignement between
word hypothesis from automatic speech recognition and manual reference)
- see header of convert/sclite.tcl file -, the acoustic segmentation .sdt
format and the word segmentation .ctm format (developed for the Coretex
European project).</li>

<li>
reworked remote playback support (the script tcl/SoundServer.tcl needs
to be launched on the client side) - feature still experimental</li>

<li>
support for client-server scripting of Transcriber through sockets.&nbsp;One
need to launch the tool with: "trans -patch tcl/Socket.tcl" - see tcl/Socket.tcl
for details on API. Experimental, intended for interface with other tools.</li>
</ul>
changes or bug fixes:
<ul>
<li>
improved support for NIST .stm format (convert/stm.tcl).</li>

<li>
store encoding used along with default settings and localization texts
for better portability.</li>

<li>
improved default setting of shape directory for Windows/Mac platforms.</li>

<li>
turn around for Tcl8.3.2 bug involving 'a grave' appearing as 'A tilde'.</li>

<li>
turn around for Tk bug resulting in a bad cursor position when moving to
the end of a word.</li>

<li>
define word boundary to be empty char on Unix platform, similar to Windows
settings.</li>

<li>
take user preference about start or end of line into account when changing
line with cursor.</li>

<li>
do not limit temporal display to 30 sec any more when the signal is not
available.</li>

<li>
case insensitive split of the apostrophe for French and Italian spell checking.</li>

<li>
selection of a speaker in the list of speakers in the turn attribute windows
now also possible with space keystroke.</li>

<li>
corrected automatic XML&nbsp;encoding detection for OS whose default encoding
is not ASCII&nbsp;compatible (e.g.&nbsp;asiatic).</li>

<li>
corrected possible display inconsistency of axis (delayed repaint).</li>

<li>
various corrections to the configuration/compilation step.</li>
</ul>

<hr width="100%"><b><font size=" 0">Version 1.4.4</font></b>
<p>Released on April 19, 2002.
<br>needs Tcl/Tk 8.0-8.3, Snack 1.7 or preferably 2.0.1, tcLex 1.2
</p><p>New or modified functions: none.
</p><p>Implementation changes or bug fixes:
</p><ul>
<li>
The cursor disappeared after playing a segment with Shift-Tab - fixed.</li>

<li>
The waveform display was sometimes not correctly refreshed - fixed.</li>
</ul>

<hr width="100%"><b><font size=" 0">Version 1.4.3</font></b>
<p>Released on June 21, 2001.
<br>needs Tcl/Tk 8.0-8.3, Snack 1.7 or preferably 2.0.1, tcLex 1.2
</p><p>New or modified functions:
</p><ul>
<li>
when Snack version 2.0 is installed with Transcriber, it is possible to
play only the left or right channel from a stereo file using the new menu
<tt>[Signal]/[Stereo
channel]</tt>. It has no effects for mono sound files or when previous
Snack version 1.7 is used.</li>

<li>
new conversion filter to .html format, which makes printing the transcription
possible from any browser.</li>

<li>
new conversion filter to .cha (Childes/CHAT) format provided by Zhibiao
Wu. Support for this format is yet experimental. When reading a ".cha"
file, the tool switches to a CHAT mode with new attributes available in
the interface. It switches back to the initial mode when creating a new
transcription or reading a file in any non-Chat format.</li>

<li>
convert filter to .stm format has been modified; import from .stm is limited
to the transcription.</li>

<li>
at first startup, the window automatically fits the whole screen; further
resizing is better managed.</li>

<li>
slight corrections for signal and cursor display.</li>

<li>
(hidden feature, activated by putting 'hideLevels&nbsp; 1' in the configuration
file): when reading a segmentation file (i.e. .lab file), hide the section/turn/background
segmentation levels under the signal since they are always empty (their
display can be switched on again using contextual menu available with the
right button on the signal)</li>
</ul>
Implementation changes or bug fixes:
<ul>
<li>
Transcriber has been ported for the Macintosh/G3 platform (thanks to L.
Spektor for the modifications). A new sub-directory 'src/mac' includes
resources for the compilation of Snack, tcLex and Transcriber on a Mac
with CodeWarrior Pro 5.</li>

<li>
automatic detection of sound files associated to a transcription file now
works also for uppercase extensions (e.g. .WAV); it used to recognize only
lowercase extensions (e.g. .wav).</li>

<li>
<tt>src/shape.c</tt> is now integrated into Snack1.7 (thanks to K. Sj&ouml;lander)
and is no longer distributed with Transcriber. Various parts of the code
(<tt>src/wavfm.c</tt>, <tt>tcl/Signal.tcl</tt>, <tt>tcl/Play.tcl</tt>,...)
has been modified according to the syntax of the new Snack commands. As
a result, Transcriber 1.4.3 doesn't work with Snack 1.6.</li>

<li>
the format of the sound shapes has changed and has now the suffix ".shape".
Previously computed shapes with ".shp" extension become useless and can
be removed.</li>

<li>
<tt>src/segmt.c</tt>: corrected a bug in BorderTableFree (attempt to free
an empty pointer when a wrong segment color name was given)</li>

<li>
<tt>src/wavfm.c</tt>: a double pixmap buffering is used for speeding up
cursor-only changes (BTW, pixmaps are now freed at the end); a new widget
sub-command 'cursor' has been created for that purpose and is used in tcl/Waveform.tcl:proc
Cursor.</li>
</ul>

<hr width="100%"><b><font size=" 0">Version 1.4.2</font></b>
<p>Released on January 6, 2000
<br>needs Tcl/Tk 8.0-8.2, Snack 1.6, tcLex 1.2
</p><p>New or modified functions:
</p><ul>
<li>
boundary move can be forced with Shift-Button2 (or Shift-Control-Button1)
- in this case, other boundaries are pushed accordingly.</li>

<li>
a break can be forced inside a segment with Shift-Return</li>

<li>
different playback modes can be chosen in [<tt>Signal</tt>]/[<tt>Playback
mode</tt>] menu (continuous, with pauses, stop or beep at segment boundaries,
or looping on segment or selection). The mode applies to any further playback.
Default pause duration, beep file, and also an initial step back before
playing are set in [<tt>Options</tt>]/[<tt>Audio file...</tt>].</li>

<li>
default position of the text cursor during automatic synchronization with
the signal can be chosen in [<tt>Options</tt>]/[<tt>General...</tt>] between
start and end of line (it used to be the end of line).</li>
</ul>
Implementation changes or bug fixes:
<ul>
<li>
in the 'Find and Replace' window, the 'Replace all' button sometimes caused
an infinite loop - replacement is now stopped at the end of the document.
Replacement of text within event descriptions failed and is now avoided.</li>

<li>
in the <tt>segmt</tt> widget, segments are now splitted vertically at first
form-feed character instead of newline.</li>

<li>
execution failed if the installation pathname contained spaces - fixed.</li>

<li>
on first launch, open dialog box in the 'demo' subdirectory rather than
in the current directory.</li>

<li>
when cancelling a font chooser and saving the configuration (i.e., [<tt>Options</tt>]/[<tt>Fonts</tt>]/[<tt>...</tt>],
then <tt>Cancel</tt>, then [<tt>Options</tt>]/[<tt>Save configuration</tt>]),
current font description was lost - fixed.</li>

<li>
always display cursor time with 3 digits precision.</li>

<li>
a date change with monitoring activated caused an infinite loop (e.g. after
a night of inactivity).</li>

<li>
under Windows, user configuration file is now saved in the $USERPROFILE
directory</li>

<li>
upon creation of a new turn, the channel setting was lost - fixed.</li>

<li>
the list of encodings is now located inside the <tt>etc/default.txt
</tt>file
for easier (manual) configuration.</li>

<li>
the character set used in any encoding remained restricted to iso-latin1
in version 1.4.1; it is now extended to the whole Unicode set for Tcl/Tk8.1
or higher. However, it is not possible to choose a font for a specific
encoding, and display of non-ascii chars sometimes fails.</li>

<li>
<tt>src/shape.c</tt> slightly modified to follow API change in Snack1.6.4</li>
</ul>

<hr width="100%"><font size=" 0"><b>Version 1.4.1 </b>(offers Unicode support
under Tcl/Tk 8.1.1 or higher)</font>
<p>Released on October 28, 1999
<br>needs Tcl/Tk 8.0-8.2, Snack 1.6, tcLex 1.2
<br>compiled under Linux/x86 (Debian 1.3/2.0), Linux/alpha, Solaris (SunOS
5.5), SGI (IRIX 6.2).
</p><p>New or modified functions:
</p><ul>
<li>
Unicode support, only available under Tcl/Tk 8.1.1 or higher. The encoding
used for writing the transcriptions can be chosen in [<tt>Options</tt>]/[<tt>General...</tt>].
It defaults to ISO-Latin-1, but can be modified to Unicode (UTF-8) or other
widespread 8-bits encodings. The same user-defined encoding is used for
reading transcriptions produced by earlier versions of Transcriber without
encoding header. Under Tcl/Tk 8.0, the behaviour remains unchanged and
makes reading transcriptions in Unicode not possible.</li>

<li>
Changing all occurences of a speaker to another already existing one is
possible by changing the old name to the new one.</li>
</ul>
Implementation changes or bug fixes:
<ul>
<li>
with Transcriber 1.4 under Tcl/Tk8.1.1 or higher, display of non-ASCII
chars in the segmentation tier under the signal was followed by garbage;
this has been corrected. Any Unicode char should be now correctly displayed,
provided it is available in an installed font.</li>

<li>
for Windows, call to <tt>XDrawLines</tt> in segmt.c with empty segment
is avoided.</li>

<li>
the transcription input/output has been modularized; modules can be added
into the <tt>convert </tt>sub-directory for supporting new input or output
formats (see <tt>convert/README</tt> file).</li>
</ul>

<hr width="100%"><font size=" 0"><b>Version 1.4 </b>(mostly a stable version
of 1.3 beta)</font>
<p>Released on September 27, 1999
<br>needs Tcl/Tk 8.0-8.2, Snack 1.6, tcLex 1.2
<br>compiled under Linux/x86 (Debian 1.3/2.0), Linux/alpha, Solaris (SunOS
5.5), SGI (IRIX 6.2).
</p><p>New or modified functions:
</p><ul>
<li>
handling of raw sound files is improved - a default header size can be
given in the [<tt>Options</tt>]/[<tt>Audio file...</tt>] window; changes
to the default raw sound files settings immediately apply to the current
sound file if necessary.</li>

<li>
background color of selected signal can be modified by the user - this
can be necessary when not enough colors are available in the colormap.</li>

<li>
a sample of speech conversation transcription is provided in the demo sub-directory.</li>

<li>
the Help is viewed through '<tt>netscape</tt>' when it is available.</li>

<li>
a -<tt>patch <i>directory_name</i></tt> option has been added to the command
line options for 'trans'; this will allow loading some extensions to the
tool at startup in the future.</li>
</ul>
Implementation changes or bug fixes:
<ul>
<li>
no more Makefile at the root directory - all compilation and installation
is done from within the '<tt>src</tt>' sub-directory.</li>

<li>
the DTD has been extended with the "elapsed_time" field in the Trans tag
(intended for total transcription time monitoring, but not currently used)</li>

<li>
shape calculation mode disabled by default for Windows platform; background
calculation is always disabled.</li>
</ul>

<hr width="100%"><b><font size=" 0">Version 1.3b7</font></b>
<p>Released on July 6, 1999
<br>needs Tcl/Tk 8.0 or 8.1, Snack 1.5p2 or 1.6a2 (do not use 1.6b1 under
Linux), tcLex 1.1.3 (do not use older release with Tcl/Tk8.1)
<br>compiled under Linux/x86 (Debian 1.3/2.0), Linux/alpha, Solaris (SunOS
5.5), SGI (IRIX 6.2), Windows NT.
</p><p>New or modified functions:
</p><ul>
<li>
check spelling (GUI for using 'ispell' external program)</li>

<li>
upon reading a transcription, look for a more recent automatically saved
version and ask user to restore from this version instead (in case of crash
during previous session)</li>

<li>
playback with pauses at segment boundaries (menu [<tt>Signal</tt>]/[<tt>Play
with pauses</tt>])</li>

<li>
significant update of French user and reference manuals</li>

<li>
remote sound file server is functional again (tcl/SoundServer.tcl has to
be launched on the host)</li>

<li>
menu for sorting an edited list (bindings, localization, etc.) upon any
field (name, value...)</li>

<li>
slight modifications in .stm format</li>
</ul>
Implementation changes or bug fixes:
<ul>
<li>
no clock calibration needed anymore (instead, native Snack "audio elapsedTime"
function available since 1.5p2 is used); cursor and callbacks management
in tcl/Play.tcl updated.</li>

<li>
mp3 sound files work with the most recent Snack versions (1.6a2).</li>

<li>
new Snack sub-command "order", to be used in conjunction with "shape" and
"get" sub-commands for remote sound file access in heterogeneous network.</li>

<li>
use a specific Snack API allowing support of Transcriber under Windows
NT platform.</li>

<li>
updated "Help" menu for display of correct .html files.</li>
</ul>

<hr width="100%"><b><font size=" 0">Version 1.3b6</font></b>
<p>Released on May 17, 1999
</p><p>New or modified functions:
</p><ul>
<li>
completely new French reference and user manual (HTML format)</li>

<li>
current cursor position in signal and vertical zoom are now saved with
session configuration</li>

<li>
at user option, saving current configuration is proposed before leaving</li>

<li>
direct access to a cursor position is now possible with menu [Signal]/[Go
to...] or by clicking on the message under the signal</li>

<li>
direct control of sound level in [Signal]/[Control Panel]; vertical zoom
moved to the same window</li>

<li>
at startup, sound files are automatically detected, even if their suffix
is unknown or lacking (in 1.3b5, they were rejected with the message "format
unknown"); more suffixes are now recognized (.aif, .aiff, .smp)</li>

<li>
move segment boundaries can also be done using control-left button 1 (instead
of middle button) for 2-buttons only mouse users.</li>

<li>
modification date and version are not editable any more in Episode window.</li>

<li>
language list is now displayed in lexicographical order of the current
langugage.</li>
</ul>
Implementation changes or bug fixes:
<ul>
<li>
corrected bug in 1.3b5 with activity logging when day changed during session
(caused infinite loop)</li>

<li>
corrected bug in update of turn and section button colors in the text editor</li>

<li>
changed duration for recalibration of clock to 10 seconds.</li>
</ul>
To be corrected or not implemented:
<ul>
<li>
"undo" very limited</li>

<li>
no spelling correction, automatic completion etc. yet</li>

<li>
no printing</li>

<li>
mp3 sound files partially working, but randomly causing Transcriber to
crash</li>
</ul>

<hr width="100%"><b>Version 1.3b5</b>
<p>Released on March 23, 1999
</p><p>New or modified functions:
</p><ul>
<li>
transcription time monitoring has been rewritten and is displayed inside
the "Informations" window if a log file is given</li>

<li>
upon startup, user is asked either for a transcription file or for an audio
file.</li>
</ul>
Implementation changes or bug fixes:
<ul>
<li>
bug in 1.3b4 with turn attributes edition</li>
</ul>

<hr width="100%"><b>Version 1.3b4</b>
<p>Released on March 16, 1999
</p><p>New or modified functions:
</p><ul>
<li>
Cut/Copy/Paste of mixed text and events works as expected</li>

<li>
Bindings replacement string can include Events (through copy/paste or directly
with selection); binding key can be typed as-is in the editor field</li>

<li>
Automatic space insertion at user option</li>

<li>
Save the geometry of some windows</li>

<li>
A user glossary is available</li>

<li>
Imported speaker can be selected; topics can also be imported the same
way. Unused topics and speakers can be removed. All these functions are
available with menu "Edit/Speakers" or "Edit/Topics"</li>
</ul>
Implementation changes or bug fixes:
<ul>
<li>
Bug correction in shape.c for 8bit wav sound files</li>

<li>
Format used in list edition since 1.3b3 slightly modified to allow a generic
n-fields edition instead of 2-fields only</li>
</ul>

<hr width="100%"><b>Version 1.3b3</b>
<p>Released on March 7, 1999
</p><p>New or modified functions:
</p><ul>
<li>
DTD: extended Event tag with attributes type (noise/lexical/pronounce/language)
and extent (begin/end/previous/next/instantaneous); previously defined
Events default to instantaneous noise.</li>

<li>
Edition of Event has been upgraded to ths new format</li>

<li>
Background shape computation is back at user option</li>

<li>
Events are displayed in the segmentation bar; they weren't any more in
1.3b2</li>

<li>
Overlapping speech is now managed inside turn window</li>

<li>
Turns and sections can be destroyed from within their edition window</li>

<li>
Information window now also gives information about transcription</li>

<li>
Signal and transcription name are both displayed in the button bar when
necessary</li>

<li>
Better localization of the interface</li>

<li>
Added menu "File/Export/..." for explicit exportation to non-native format;
"File/Save" and "File/Save as..." always use native XML format</li>

<li>
Default extension of transcription files has been changed to .trs; .xml
can still be used</li>

<li>
User configuration of:

<ul>
<li>
language list (in iso639 code)</li>

<li>
noise events list</li>

<li>
pronounce events list</li>

<li>
lexical events list</li>

<li>
display format of events, their color</li>

<li>
bindings</li>

<li>
localization strings in any language</li>
</ul></li>
</ul>
Implementation changes or bug fixes:
<ul>
<li>
Switched to Snack1.5 (final), tcLex1.1 (final) and tcl8.0.5 for development
(should still work with previous 8.0.x)</li>

<li>
C sources: reworked shape.c for use of latest Snack function and shape
display of any type of sound file, including compressed ones.</li>

<li>
File format .typ has been corrected to come back to 1.2 behaviour</li>

<li>
Updated Transcriber URL to its new home page inside distribution and docs</li>

<li>
no more patch to tk_getSaveFile lib for choosing the save format</li>
</ul>

<hr width="100%"><b>Version 1.3b2</b>
<p>(not publically) released on February 12, 1999
</p><p>New or modified functions:
</p><ul>
<li>
DTD:

<ul>
<li>
added channel attibutes in Turn tag (=telephone/studio)</li>

<li>
added scope attribute in Speaker tag (=local/global)</li>
</ul></li>

<li>
Added "File/Import speakers..." menu: import all speakers from a .xml file</li>

<li>
Automatic insertion of spaces before punctuations</li>

<li>
Suppressed "File/Close" menu</li>

<li>
Write to .stm format for further exploitation</li>
</ul>
Implementation changes or bug fixes:
<ul>
<li>
Modifs to Xml parsing libs</li>
</ul>

<hr width="100%"><b>Version 1.3b1</b>
<p>Released on January 21, 1999 - test under Linux (Debian 1.3/2.0), Solaris
(SunOS 5.5), SGI (IRIX 6.2).
</p><p>New or modified functions:
</p><ul>
<li>
a new XML format becomes the native format for the tool instead of LDC's
.typ</li>

<li>
validity of the transcription is checked upon the DTD (i.e. the syntax
of the format)</li>

<li>
general informations about the transcriber name, the emission or the signal
file are now kept along with the transcription</li>

<li>
when opening a transcription, the corresponding sound file is open automatically</li>

<li>
background music or noise is now handled and displayed in a separate transcription
tier</li>

<li>
Speech or non-speech events and of comments are now handled as XML tags
rather than normal text</li>

<li>
management of overlapping speech has been reworked</li>

<li>
management of speakers and topics is easier and more powerful, and allows
global research and edition of speakers or topics</li>

<li>
automatic shape computation for large signals can be disabled at user option</li>

<li>
loading raw files is once again a problem - but it should work even better
than before with future updates of Snack module...</li>

<li>
transcriber's productivity can be logged in a file at user option</li>
</ul>
Implementation changes:
<ul>
<li>
SoX libraries aren't used anymore; instead, all sound files are accessed
with Snack module; specific open_sound and play_sound command disapeared</li>

<li>
tcLex module from Fr&eacute;d&eacute;ric Bonnet (a lexical analyser generator
for Tcl) is now used for the implementation of an object-oriented tcl-only
validating XML parser</li>

<li>
installation procedure and global tool structure has been reworked thanks
to D. van Leeuven useful hints</li>
</ul>

<hr width="100%"><b>Version 1.2</b>
<p>Released on September 16, 1998 - test under Linux (Debian 1.3 and Redhat
4.3), Solaris, SGI (IRIX 6.2).
</p><p>New functions:
</p><ul>
<li>
Signal:

<ul>
<li>
Display topic and speaker segmentation under the waveform along with text
segmentation.</li>

<li>
Ability to display a second view of the signal at a different scale.</li>

<li>
Possibility to open raw sound files.</li>

<li>
Contextual menus on the waveform for direct actions on the signal and its
display.</li>
</ul></li>

<li>
Transcription:

<ul>
<li>
Undo/Redo, limited to segmentation boundary moves and text changes (undo
all text modifs in the current segment since cursor moved into it).</li>

<li>
Find and replace in the text.</li>

<li>
Automatic auto-save at periodic intervals.</li>

<li>
New XML format for transcription output (format will probably be modified
soon - see discussion about formats).</li>
</ul></li>

<li>
Management of overlapping speech over one segment.</li>

<li>
More configurable options (fonts, colors, windows to display...). Most
user choices (including current sound file, transcription file and current
resolution display) are saved in the "~/.transcriber" user configuration
file upon request.</li>

<li>
Localization of the menus in english or in french as option; other languages
can be easily added.</li>

<li>
Help in HTML format.</li>
</ul>
Bug fixes:
<ul>
<li>
Topics were lost when file was saved as ".typ". Original LDC format was
slightly modified to include topic. Markup for overlapping speech is now
recognized.</li>

<li>
Default configuration file was named "~/.trancriber" (without 's'); it
was corrected.</li>

<li>
Shortcut for 'Signal/Replay Segment' has been moved from &lt;Alt-Tab&gt; to
&lt;Shift-Tab&gt;; it could be a problem on some platforms.</li>

<li>
Menus 'Segmentation/Move to.../First segment' and '.../Last segment' are
now executed.</li>

<li>
... plus a lot of minor changes</li>
</ul>
Implementation changes:
<ul>
<li>
Internal data management has been completely rewritten. Transcriptions
are stored in memory as a tree structured like an XML file, and are accessed
with a set of procedures in "tcl/Tags.tcl". Input file parsing is very
simplified (an XML-conformant library using tcLex extension for Tcl written
by Fr&eacute;d&eacute;ric Bonnet is under development).</li>

<li>
Sound files are accessed via a new Tcl command "open_sound" which calls
sox and/or sphere libraries and creates a new "sound-file" Tcl command
- not (yet) compatible with Snack "sound" commands. Playback with a new
Tcl command "play_sound" which calls Snack libraries. These functions are
documented in the "doc" sub-directory.</li>

<li>
Sound shape (pre-computed at the resolution of 10 ms per pixel) are not
computed in background any more. Rather, the interface is stopped the first
time a large sound file is open during the shape computation; it is then
saved for the next time it is needed.</li>

<li>
Remote audio playback server is no longer available; it seemed not to be
essential, and Snack playback design is more robust. Remote audio file
server is still in the distribution, but the default behaviour is to read
audio files locally without launching the audio file server on the local
host. If needed, the file server should be configured before launching
to avoid security holes (e.g. by testing the client port or the sound file
directory).</li>

<li>
axis, wavfm and segmt new Tk widgets have been updated (but they are not
yet documented):

<ul>
<li>
wavfm widget now calls a "sound-file" command created with the "open_sound"
command for shape calculation rather than reading/writing a channel (pipe
or socket). Widget options "-width" and "-height" added.</li>

<li>
segmt widget support one indirection level for time indices : start and
end segment times can be an index in an associative array. Time array and
list of segment variables are automatically traced, and the widget is redisplayed
if they are modified. Some minor other fixes.</li>

<li>
axis and segmt widgets are redrawn if named fonts are modified (allows
interactive fonts configuration).</li>
</ul></li>
</ul>

<ul>
<li>
Distribution structure:

<ul>
<li>
the main distribution directory is now named Trans-x.y (x.y being 1.2 in
this version). For simplification, the "editor" sub-directory has been
suppressed and its content ("doc", "img", "src", "tcl") is in the first
level. The "pkg" sub-directory has been renamed "lib". The "bin" and "obj"
sub-directory are no longer used.</li>

<li>
all C extensions to Tcl/Tk are grouped in a single auto-loadable package
"libtrans.so"; there are no binary executable to be launched in background
any more.</li>

<li>
a patched version of the last distribution of SoX (v-12.14) is included
in the distribution in the "contrib" sub-directory</li>

<li>
html_library-0.3 from Stephen Uhler is included in the "lib" sub-directory.</li>

<li>
Snack official distribution is used without any patch. The lite version
of the package is installed in the "lib" sub-directory.</li>

<li>
a configure procedure is available for the compilation of C sources in
the "src" sub-directory</li>
</ul></li>
</ul>

<hr width="100%"><b>Version 1.1</b>
<p>Released on June 26, 1998 - test under Linux (Debian), Solaris, SGI.
</p><p>Modifications:
</p><ul>
<li>
SNACK 1.1 module from K&aring;re Sj&ouml;lander is used instead of Sox
libraries for audio playback. This solution is more stable and portable
: Snack works under HP-UX, Sun Solaris, SGI IRIX, Linux, and Windows 95/NT.
The module is used in a patched version.</li>

<li>
No modifications of the user interface, installation procedure still rather
crude.</li>
</ul>

<hr width="100%"><b>Version 1.0</b>
<p>Released on May 25, 1998 - development under PC/Linux (Debian 1.3);
test under Sun Solaris.
</p></body></html>