Tips'n'Scripts

I use the txt2pho program to convert the text into phonemes, and mbrola for transforming the phonemes into spoken audio. I do some extra preprocessing of the text to fit my needs:

#!/bin/sh
if [ x"$#" = x"0" ]; then
  file=/dev/stdin
  else
        file=${1}
fi
cat $file \
| sed 's/\.com/ dot comm /g' \
| sed -r 's/00:([0-9]+)/0 Uhr \1/g' \
| sed -r 's/([0-9]+):00/genau \1 Uhr/g' \
| /usr/share/txt2pho/preproc/preproc /usr/share/txt2pho/preproc/Rules.lst /usr/local/txt2pho/preproc/Hadifix.abk \
| sed 's/cweiske/ c weiske /g' \
| /usr/bin/txt2pho -m \
| /home/cweiske/scripts/say-de5 \
| aplay -

You can pipe your text into the program, e.g. via echo "hello you there" | say or just give a filename which is read. Then there are some conversions to my liking; adresses ending with ".com" are pronounced "dot com" while german addresses ".de" are pronounced "Punkt de", which is the german speak. Then some time conversion is done as the normal is suboptimal. And of course "cweiske" is spoken "c weiske".

The say-de5 is the script for converting the phonemes into audio with the german voice number 5. For fast changing I put this in an extra script so that the pitch and loudness can be specific for every voice.

#!/bin/sh
#german female voice
cat /dev/stdin \
| /usr/bin/mbrola -v3 -f2 -t1 /usr/share/txt2pho/voices/german/de5 - -.au

Now I made another script which just passes all the parameters to the say script, as some programs just can call one program with the text as the parameter (like Opera)

#!/bin/sh
echo $* | say

Last but not least there is a script which tells the time. I put it in a cron job which is called every ten minutes.

#!/bin/sh
echo Es ist jetzt `date '+%H:%M'` | say

Here is a description of how I integrated the tts scripts into some programs for speech output

...
[Hotclick Popup Menu]
Item, 50872 = Copy
Item, "Say" = "Copy & Execute program,"/home/cweiske/scripts/sayparam","%c"" 
Item, 67652 = Copy to note
...

echo "%t" | say

#!/bin/sh
# emerge it
emerge $*

#now say it
#we add a space before so that we can be sure to have a standalone -f parameter
allparams=" $* "

if [ `echo "$allparams" | grep -e ' -f ' | wc -l` = 1 ]; then
    #all downloaded
    text="Es wurde alles herunter geladen"
elif [ $# -eq 1 ] && [ $allparams = "sync" ]; then
    #sync complete
    text="Synchronisation abgeschlossen"
elif [ $1 = "search" ]; then
    #search complete, tell the number of found applications
    number=`emerge $* 2> /dev/null | grep "Applications found :" | sed 's/^.*: //' | sed 's/ ]//'`
    number=`echo $number | sed s/^0$/keine/`
    text="Bei der Suche wurden $number Programme gefunden"
else
    #normal installation
    text="Installation abgeschlossen"
fi
echo "$text" | /home/cweiske/scripts/say &> /dev/null &

My favourite email program.

When I call thunderbird while there is an open instance, a new instance is started and the "select profile" dialog is shown. I don't like this behaviour, and so here is a script which just activates a running instance if there is one.

#!/bin/sh
sudo /scripts/mail-popfile

if [ `ps -u cweiske | grep thunderbird-bin | wc -l` -eq 0 ]; then
        #echo "start thunderbird new"
        /usr/bin/thunderbird &
else
        #echo "thunderbird is running, just activating"
        thunderbird -remote "xfeDoCommand(openInbox)"
fi

The mail-popfile scripts takes care that stunnel (ssh program) and popfile (spam filter) is started, so that I can filter my mails with popfile and use a SSL connection to the mail server at the same time.

#!/bin/sh

#stunnel - already running?
numstunnel=`ps uax | grep stunnel | wc -l`
if [ $numstunnel -ge 2 ]; then
        echo "stunnel is running"
else
        echo "starting stunnel"
        /usr/sbin/stunnel
fi

#popfile
numpopfile=`ps uax | grep popfile | grep perl | wc -l`
if [ $numpopfile -ge 1 ]; then
        echo "popfile is running"
else
        echo "starting popfile"
        cd /usr/share/popfile
        ./popfile.pl &
fi

echo "mail-popfile done"

Normally firefox is started when there is a link in an email in thunderbird, but I am addicted to Opera... So I changed the http protocol handler that it starts Opera now

...
user_pref("network.protocol-handler.app.http", "/home/cweiske/scripts/browser");
...

with the browser script being the following

#!/bin/sh
# opens opera with the given url

url="$1"
if [ "x$url" = "x" ]; then
        url="about:blank"
fi

exec opera -newpage "$url"

Now it is comfortable with thunderbird, but I am still missing some things:

To call thunderbird on mailto: Links, I set the the email program setting to "Use specific e-mail client" which is set to /home/cweiske/scripts/tb-compose

#!/bin/sh
# dont' rename it to "thunderbird-compose" or so
# cause the thunderbird will be found in the process list

if pgrep -u `whoami` thunderbird
then
        /opt/MozillaThunderbird/thunderbird -remote "mailto($1)"
else
        /opt/MozillaThunderbird/thunderbird -compose mailto:$1
fi

Here is the predefined subject and text missing, but that's got to be made another time...

#!/bin/sh
#check if kaffeine is running
kaffeine=`dcop | grep kaffeine -c`
if [ $kaffeine = '1' ]; then
    #kaffeine is running, control it
    dcop kaffeine Kaffeine togglePlayPause
elif [ `pgrep tvtime` ]
then
    #tvtime
    tvtime-command TOGGLE_PAUSE
    tvtime-command TOGGLE_MUTE
else
    #no kaffeine, no tvtime
    #xmms play-pause
   xmms -t
fi

This code does the following: - if Kaffeine (KDE video player) is running, a play/pause signal is sent to it - if tvtime (TV application) is running (but not kaffeine), the picture is paused and the sound is muted in tvtime - if none of the above, XMMS shall play/pause.

Main problem was that the notebook disk is only 30 gigabytes, and there were only 4 of them free. The space was filled up after 15 minutes digital video were transferred... As the desktop pc had about 50 gig free, I tried to mount the free drive via nfs on the laptop and directly grab the files over network. Unfortunately the connection wasn't fast enough (100mbit network), but I had dropped frames every now and then. So I grabbed to the laptop disk and moved the files with another console to the nfs drive. No problems with dropped frames.

I wanted to use the desktop pc for copying the other two tapes to disk, so I bought a FireWire card and compiled the FireWire modules into the kernel. After a restart I could open Kino and control the camcorder - but I didn't see any video. dvgrab didn't give any results either, although gscanbus allowed me to control (play/pause/stop) the camera, too. Searching the net brought up another user with the same problem, but no solution. I subscribed to the FireWire mailing list at linux1394.org and told them my problem. I had an answer in my mailbox the next morning telling that some users got it working if they removed the FireWire ethernet driver eth1394. And after re-compiling the firewire kernel drivers as modules it worked as long as eth1394 was not loaded!

After copying one hour of video (16 gigabytes of disk space used!), I used Kino to arrange the files and cut the scenes down to usable ones. As I used dvgrab --autosplit --timestamp for grabbing the video from the digital camcorder, I had about 200 single files to import into Kino. As Kino doesn't support opening multiple files, dragging them from Konqueror helped to get them at once. Now there was no button to delete scenes, so I used Trim to set the beginning and the end mark to the same position. After doing that for about 100 of the 200 scenes, I noted that one could use the Cut command to get them out. But the program is somewhat slow cutting them (as it recreated the index or somewhat), I closed it and tried to do that by hand: The project files are in SMIL format, an XML dialect for audio and video sceneries. So it's human readable, and I opened it in my favorite editor and cut out all the lines which had the same beginning and ending position. That was done after one minute, compared to the four scenes you can cut with Kino in that time.

The next problem was getting rid of all the files I didn't need any more because I cut them out: The shell was my friend in that case.

At first I needed a list with all files that were in use:

fgrep src kinoproject.smil | cut -b 17-63 > files_video

wrote all the filenames used in the project file into files_video. It could be that the "63" is different for you if the dvgrab filenames have another length.

The next was to get a list with all files not in the video:

ls -1 video/ | grep -v -f files_video > files_to_move

That one lists the files in video/ directory, one each line and compares them with the lines in the file created previously. The result is a list with all files not in the project.

Now all I had to do was moving the files to another directory:

for i in `cat files_to_move`; do mv video/$i /mnt/otherdrive/; done

You could delete the files here, too.