[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Fwd: Re: request for emacspeak installation instructions on mandrake 7.2



Hi Dave,
It took me several tries to get emacspeak working with Mandrake 8.0, but it 
finally happened. I don't have audio icons, and viavoice still crashes 
periodically in C/C++ mode, but at least the install procedure seemed to work.

The following are the notes that I made so that I could repeat the process.

I hope this helps. If anyone else wants to fix my notes, feel free. If you 
find errors, please let me know, so I can correct my notes.

Good luck,
Marty





At 09:02
AM 10/11/01 -0700, you wrote:

>>Roger,
>
>
>If your experience with installing emacspeak onto a Mandrake installation 
>is universal, then I for one am in the deep unplesant stuff.
>
>I've had considerable trouble getting emacspeak installed with via voice 
>and still do not have it working.  I've had to resort to locating somebody 
>with experience at this stuff.  Funny thing is, that Mandrake installed 
>very easily -- very smooth, no hitches.  Getting bootstrapped with 
>emacspeak has been just the opposite.  I'm getting a certain degree of the 
>depth of my aversion to Windows, by the fact that I'm still pursuing linux 
>as the desired alternative.
>
>So, are there others who actually have emacspeak 14.0 with via voice 
>running under Mandrake 8.0??? If not, I've wasted $80!!
>
>Dave
>
>
>
>>From: Roger <roger@cclawyer.com>
>>To: Aruni <aruni_sharma@yahoo.com>, emacspeak@cs.vassar.edu
>>Subject: Re: request for emacspeak installation instructions on mandrake 7.2
>>Date: Thu, 11 Oct 2001 09:41:14 -0500
>>
>>Hi Aruni,
>>
>>I've found that Mandrake is so install friendly, that it writes alot of stuff
>>that you don't need and have to hunt, find, eliminate, just to get emacspeak
>>and viavoice to work.
>>You really need RH, thats what most of the people on the list use, I believe.
>>And its the only one I have been able to get to work, though I have tried
>>Mandrake, SuSe, and the kitchen sink. If ViaVoice wasn't so proprietary, I
>>think the BLINUX project would already be bearing fruit with a simple single
>>install (and of course some tweaking).
>>
>>  If you or a friend has DSL and a CD burner you can go to the following link
>>and download ISO images of most all the distributions, including RH 7.1.  The
>>download will still take 3-5 hours per disk, and there are two disks,
>>minimum, plus an optional power-tools disk.
>>  http://www.linuxiso.org/
>>
>>  The Gnome Java stuff looks really promising!! I hope KDE follows suit as
>>well.
>>
>>Peace In Troubled Times,
>>rog
>>
>>
>>
>>
>>
>>
>>On Saturday 06 October 2001 11:02 am, Aruni wrote:
>> > Hello, everyone. I've just got mandrake 7.2 installed on my system.
>> > However, I don't have any idea about Linux. So will somebody help me with
>> > specific instructions on how to install viavoice and emacspeak on mandrake
>> > 7.2. Actually I was looking for Redhat7.1. But I could not get hold of it.
>> > thanks in advance,
>> > Aruni.
>> >
>> >
>> > _________________________________________________________
>> > Do You Yahoo!?
>> > Get your free @yahoo.com address at http://mail.yahoo.com
>> >
>> > 
>> ---------------------------------------------------------------------------
>> >-- To unsubscribe from the emacspeak list or change your address on the
>> > emacspeak list send mail to "emacspeak-request@cs.vassar.edu" with a
>> > subject of "unsubscribe" or "help"
>>
>>-----------------------------------------------------------------------------
>>To unsubscribe from the emacspeak list or change your address on the
>>emacspeak list send mail to "emacspeak-request@cs.vassar.edu" with a
>>subject of "unsubscribe" or "help"
>
>e-mail: helkenn@titan.com
>Voice: (858) 552-9916    FAX: (858) 552-9660
>s-mail: Titan Linkabit  3033 Science Park Rd.  San Diego, Ca 92121
>
>-----------------------------------------------------------------------------
>To unsubscribe from the emacspeak list or change your address on the
>emacspeak list send mail to "emacspeak-request@cs.vassar.edu" with a
>subject of "unsubscribe" or "help"
Installation instructions for Emacspeak:

Ehese instructions reflect the installation of Emacspeak-14 on a Mandrake
8 Linux system, tunning on an Dell Optiplex GX-1 desktop.

1. Download/copy and install Tcl, Tclx, Tk:

   a. rpm -ivh tcl-8.3.2-7mdk

   b. rpm -ivh tk-8.3.2-7mdk	

   c. rpm -ivh tclx-8.3.2-7mdk

2. Downloaad/copy and install the Lesstif packages (Lesstif-mwm, Lesstif-devel, andLesstif-client) which are available from www.lesstif.org. Use the following rpm commands to install the packages:

    rpm -ivh lesstif-mwm-0.92.32-1.i386.rpm
    rpm -ivh lesstif-clients-0.92.32-1.i386.rpm
    rpm -ivh lesstif-devel-0.92.32-1.i386.rpm

a. If you get a failed dependencies error for LibXm.so.2 and/or LibMrm.so.2, don't worry. Niether of these packages are required for the command line version of emacspeak. You should modifiy your rpm commands to:

	 rpm -ivh --nodeps Lesstif-mwm-0.92.32-1.i386.rpm
	 rpm -ivh --nodeps Lesstif-clients-0.92.32-1.i386.rpm
	 rpm -ivh --nodeps Lesstif-devel-0.92.32-1.i386.rpm

b. If  you get another error, check it out prior to continuing.
	 
3. Download/copy osslinux395c-2x.tar.gz for www.leb.net. This is the free download version of the OSS sound drivers. Unpack the package with the following:

   gunzip  osslinux395c-2x.tar.gz
   tar -xvf osslinux395c-2x.tar


4. Download oss into /tmp

   ./oss-install

a. the oss-install program will detect existing sound drivers. If drivers exist, unload it by executing: 

   rmmod sound

b. Try ./oss-install again. If error repeats, you probably have the driver being loaded by the kernel daemon. Remove all sound.o files from all subdirectories of /lib/modules.conf and remove the "alias sound ..." line and replace it with "alias sound off". Also add a line containing "alias char-major-14 off" to /etc/modules.conf. The program will offer to make the changes automatically.

5. Configure the sound card by running:

   sndconfig

6. Download, unpack, and install stdiom:

   gunzip stdiom.tar.gz
   tar -xvf stdiom.tar

   cd stdio_musician1.0
   make


7. Download/copy, unpack, and install viavoice_tts_rtk and viavoice_tts_sdk:

a. Viavoice_tts_rtk_5.tar INSTALLATION INSTRUCTIONS

   tar -xvf viavoice_tts_rtk_5.tar
   
1)  Log in as root.

2)  If you are installing the files from your drive, at a command line, enter 
    the following commands:
     
         # rpm -ivh ViaVoice_TTS_rtk-5.1-1.1.rpm

    If you are installing directly from the CD, at a command line, enter 
    the following commands:

         # mount /mnt/cdrom 

         # rpm -ivh /mnt/cdrom/<path>/ViaVoice_TTS_rtk-5.1-1.1.rpm 

         # umount /mnt/cdrom

Note: The <path> is the location of ViaVoice rpm file on cdrom.

Note: If you are missing dependencies the software can still be installed by 
specifying the --nodeps option on the rpm command line.  In doing so, however, 
some functionality may be missing.


The RPM install modifies the /etc/profile. You must either log off and back on,or run:

    source /etc/profile

The RPM install executes an initialization program named inigen. This program 
creates a file named eci.ini. When you write a program that uses speech 
synthesis, you must link to our libibmeci50.so library. When libibmeci50 runs, 
it looks in the environment to see if the string ECIINI is defined. If so, it 
uses the value of this string to find eci.ini. If not, it looks in the current 
directory. It needs eci.ini in order to determine what synthesis languages are 
installed on your machine, and where they are. You can create your own eci.ini 
by running inigen yourself. We put inigen in /usr/lib/ViaVoiceTTS/bin. It 
takes two arguments:
    inigen <full path and filename of language engine> <full path to eci.ini>

   Example:
   /usr/lib/ViaVoiceTTS/bin/inigen /usr/lib/enu50.so       
   /usr/lib/ViaVoiceTTS

Also, make sure no other programs are using the audio devices on your system.


8. Before installing Viavoice_tts_sdk, you should determine if your audio is working before you start the install. If your audio is not working properly, check your hardware. To test the audio, play a wav file using the following:

          # cat /usr/share/sounds/ktalkd.wav > /dev/audio
          (This will play the ktalkd.wav file.) 

  If you don't have ktalkd.wav file on your system, play an alternate wav file 
  of your choice by substituting the file name in the above command.

  If the above audio/playback test does not work, verify your hardware 
  and Linux operating system are properly configured. Hint: Use sndconfig 
  to configure the sound card in your system.

b. Installing the Viavoice Software devel kit. Unpack with:

   tar -xvf viavoice_tts_sdk_5.tar
 
1)  Log in as root.

2)  If you are installing the files from your drive, at a command line, enter 
    the following commands:
     
   rpm -ivh ViaVoice_TTS_sdk-5.1-1.2.i386.rpm

If you get a dependencies error for LibXm, use this instead:

   rpm -ivh --nodeps ViaVoice_TTS_sdk-5.1-1.2.i386.rpm

Note: If you are missing dependencies the software can still be installed by 
specifying the --nodeps option on the rpm command line.  In doing so, however, 
some functionality may be missing.

Make sure no other programs are using the audio devices on your system.
ViaVoice TTS SDK for Linux includes the following three sample programs: 

1) Command Line Speak Sample
2) File Speak Sample
3) Lesstif Speak Sample

To run the Command Line Speak Sample program:

   cd /usr/lib/ViaVoiceTTS/samples/cmdlinespeak

   ./runcmdlinespeak

When the program is executed, it will speak the words shown:

       "Hello, welcome to ViaVoice TTS for Linux."

To run the File Speak Sample program:

   cd /usr/lib/ViaVoiceTTS/samples/cmdlinespeak

   ./runcmdlinespeakfile

When the program is executed, it will speak the words in the file:

  /usr/lib/ViaVoiceTTS/samples/cmdlinespeak/american-voices-eci.txt

This file contains samples of text annotation, which is how the synthesized 
speech can be changed to emphasize words, display emotions, or change voice 
attributes.

9. Download/copy, and install Emacspeak:

   rpm -ivh emacspeak-14.0-1.noarch.rpm

Build and install support for the speech server. Change to the Linux-outloud subdirectory:

      cd /usr/share/emacs/site-lisp/emacspeak/servers/linux-outloud

      make

This will compile the Emacspeak Outloud TCL library to create a file called tcleci.so.

      make install

This should copy over the relevant files to where Emacspeak has been installed.

10. Testing the Emacspeak-Outloud support. The outloud server is located in the emacspeak/servers directory.?????????


11. Start Emacspeak, citing the outloud speech server:

    emacspeak -o

12. Modify the configuration file 

Emacspeak Files | Subscribe | Unsubscribe | Search