Visit Schott’s German IPA FLAC files (section: deep) or Voxforge. Download and import the German speech model ‘deep’.
Archive for the ‘import’ Category
German speech model ‘deep’
Sunday, August 7th, 2011German speech model ‘deek’
Wednesday, August 3rd, 2011Visit Schott’s German IPA FLAC files (section: deek) [source 1] or Voxforge [source 2]. Get the corresponding speech model [object], and import it into simon 0.3.
German speech model ‘dedg’
Sunday, March 27th, 2011You can import the German speech model ‘dedg’ into simon. Find the corresponding FLAC files at section dedg or at Voxforge. About 10% of the following words were recognized correctly:
Lohnsteuerjahresausgleich Lohnsumme Lutschtablette Lombardkredite Luftstrahlsiebung Luftfrachtraum Luftfrachtsendung Luftfrachttarif Lufthansa-Maschine Lufthansas Lufthebebohranlage Ludendorffs Lukenwinde Lutschbonbon Lotschnur Lokschuppen Luftsprung Luferin Lords Lostrennen Lnderfinanzausgleich
You can see that the German special characters äöüß are missing. I don’t know how to solve this issue. When I type into the keyboard, everything is fine with äöüß.
German speech model ‘dedb’
Sunday, March 27th, 2011You can import the German speech model ‘dedb’ into simon. The words are from section dedb. Find them at Voxforge.
German speech model ‘decw’
Saturday, March 26th, 2011You can import the German speech model ‘decw’ into simon. The words are taken from section decw, and they can be found at Voxforge, too.
German speech model ‘decr’
Friday, March 25th, 2011You can import the German speech model ‘decr’ into simon. The words are from section decr; you can find them at Voxforge, too.
German speech model ‘dech’
Friday, March 25th, 2011You can import the German speech model ‘dech’ into simon. The words are from section dech, and they can be found at Voxforge, too.
Latin speech model ‘xcs’
Friday, March 25th, 2011You can import the Latin speech model ‘xcs’ into simon. The words are from section xcs. They can be found at Voxforge, too. The following words were recognized (50% correctly):
coemamus comamini comatum combinabimur combinaturos combinaturas combinetur comedendos comedentibus comederer comedimur comedunto commacularet commacularit commaculasset commaculatas commeabant commeabimur commeamini commeandis commereatur
Latin speech model ‘xcn’
Friday, March 25th, 2011You can import the Latin speech model ‘xcn’ into simon. It contains the words from section xcn which are available at Voxforge, too. The following words were recognized with a recognition rate of about 40%:
caestui caeptorum caeptos ceperitis certata certatis certatote certaturis certaveratis certavimus certem cessabat cessabunt cessanti cessantis cessaremur cessaris cessatura cessaturum cessaveris cessero cessisti cessuro crepaberis crepabuntur crepaturas crepida crepidinum crepitarimus crepitaro
Latin speech model ‘xci’
Thursday, March 24th, 2011You can now import the Latin speech model ‘xci’ into simon. The words are from section xci. You can get the corresponding FLAC files from Voxforge. About 50% of the following words were recognized correctly:
hiscite histricae historiis histrionis homicidiis hostilium Homeros hiscendarum honestatibus honorabunt honorans honorarimus honoratur honestissimorum honoraveram horoscopanda horoscoparetur hortatuum Hirtii hostia hispidae hostiam hosticorum humanitas humanitate humanitatem humanitates humanitatis humanitatum
Latin speech model ‘xcd’
Thursday, March 24th, 2011You can import the Latin speech model ‘xcd’ into simon. You can find the words (section xcd) of this speech model at Voxforge.
Latin speech model ‘xbt’
Tuesday, March 22nd, 2011You can import the Latin speech model ‘xbt’ into simon. It contains words from section xbt. The words can be found at Voxforge, too.
Latin speech model ‘xbo’
Monday, March 21st, 2011You can import the Latin speech model ‘xbo’ into simon. The words are from section xbo, and they can be found at Voxforge, too.
Latin speech model ‘xbj’
Monday, March 21st, 2011You can download the Latin speech model ‘xbj’ (and of course use it with simon). It contains words from section xbj. The source audio files can be downloaded from Voxforge, too.
Latin speech model ‘xbe’
Monday, March 21st, 2011How I create the Latin speech model ‘xbe’:
1. Make directory via Ubuntu terminal:
mkdir /media/104d991d-2062-40d7-89f6-ddde3cb5b781/home/ubuntu/Documents/2011-i/latin-0.1.3/split/xbe-folder/latin-speech-model-xbe
2. Copy hmmdefs:
cp /tmp/kde-ubuntu/simond/default/compile/hmm24/hmmdefs /media/104d991d-2062-40d7-89f6-ddde3cb5b781/home/ubuntu/Documents/2011-i/latin-0.1.3/split/xbe-folder/latin-speech-model-xbe/hmmdefs-xbe
3. Copy macros:
cp /tmp/kde-ubuntu/simond/default/compile/hmm24/macros /media/104d991d-2062-40d7-89f6-ddde3cb5b781/home/ubuntu/Documents/2011-i/latin-0.1.3/split/xbe-folder/latin-speech-model-xbe/macros-xbe
4. Copy stats:
cp /tmp/kde-ubuntu/simond/default/compile/stats /media/104d991d-2062-40d7-89f6-ddde3cb5b781/home/ubuntu/Documents/2011-i/latin-0.1.3/split/xbe-folder/latin-speech-model-xbe/stats-xbe
5. Copy tiedlist:
cp /tmp/kde-ubuntu/simond/default/compile/tiedlist /media/104d991d-2062-40d7-89f6-ddde3cb5b781/home/ubuntu/Documents/2011-i/latin-0.1.3/split/xbe-folder/latin-speech-model-xbe/tiedlist-xbe
6. Get GPL license text:
wget http://script.blau.in/etc/GPL_License
7. You can import the speech model into simon (Manage scenarios and Static model). The source audio files are taken from section xbe, and can be found at Voxforge, too.
How to compile a speech model
Friday, March 4th, 2011This article explains how you can compile the German speech model ‘decc’:
1. Get the package decc from Voxforge. Download the file german-ipa-flac-files-decc-20110302.tar.bz2, and extract it. The extracted folder has the location on my computer: file:///media/104d991d-2062-40d7-89f6-ddde3cb5b781/home/ubuntu/Documents/2011-i/german-0.2.5/object/split/decc/german-ipa-flac-files-decc-20110302
2. Remove old directories via Ubuntu terminal:
rm --recursive /home/ubuntu/.kde/share/apps/simon
rm --recursive /home/ubuntu/.kde/share/apps/simond
3. Quit simon. Quit ksimond. Start simon.
4. simon > Manage scenarios > New > Name: de-cc
Press the Add button, and enter a Name and a Contact. Here is my solution:
Name: http://script.blau.in/german/decc.xml
Contact: 2011
Press OK.
5. Move the scenario de-cc from the Available column to the Selected column. Press OK.
6. simon > Vocabulary > Wordlist > Active vocabulary > Import dictionary
7. Target: Active dictionary.
8. Select the type of dictionary > PLS lexicon
9. Set the path to the dictionary lexicon-decc.xml
10. Grammar > Grammar > Add sentence
11. Enter the Grammar structure. You will need three structures: Adjektiv, Substantiv and Verb.
12. Training > Training > Import training data
13. Import prompts. You have to set the path to the prompts file prompts-decc. And set the path to the prompts directory flac-decc.
14. Note: It is assumed that the following condition is met: Recordings > Post-processing > "sox -t flac %1 -t wav % 2" Otherwise, step 13 would have failed. Thanks to this instruction, simon will convert the FLAC files from FLAC to WAV during import.
15. Commands > Commands > Manage plug-ins
16. Select the dictation plug-in from the list. Append text after result: " " (press the space bar one time so that simon will insert a space character after each word)
17. Start ksimond. simon > Press the Connect button. Wait a few moments because simon is now compiling the speech model.
18. simon activates itself automatically. This means now is the time to dictate a few words:
Golfspielers Golfprofi Golfprofis Golfrasen Golfspielen Gondelhandel Gottbekenntnis Gottbekenntnisse Gottbekenntnissen Gottbekenntnisses Gottesfurcht Gottesdiensten Gottesdienste Gottesglauben Gradualismus Gradmesser Gradienten Graphologe Grammatom Grammatur Granitblock Granitblocks Graphensprache Grashalmes Gratulationen Gratulationsempfang Gratulationsschreiben
Most words were recognized correctly.
19. Now you know how to compile a speech model. But wait, where are the files hmmdefs, tiedlist, macros, stats? On my computer, these are the paths:
file:///tmp/kde-ubuntu/simond/default/compile/hmm24/hmmdefs
file:///tmp/kde-ubuntu/simond/default/compile/hmm24/macros
file:///tmp/kde-ubuntu/simond/default/compile/stats
file:///tmp/kde-ubuntu/simond/default/compile/tiedlist
These files will be included in the German speech model ‘decc’. The corresponding scenario can be exported (simon > Manage scenarios > Export).
20. You can download the German speech model ‘decc’, and import it as static model.
21. Here is how I create this speech model (via Ubuntu terminal):
a. Make directory:
mkdir /media/104d991d-2062-40d7-89f6-ddde3cb5b781/home/ubuntu/Documents/2011-i/german-0.2.5/object/split/decc/german-speech-model-decc
b. Copy hmmdefs:
cp /tmp/kde-ubuntu/simond/default/compile/hmm24/hmmdefs /media/104d991d-2062-40d7-89f6-ddde3cb5b781/home/ubuntu/Documents/2011-i/german-0.2.5/object/split/decc/german-speech-model-decc/hmmdefs-decc
c. Copy macros:
cp /tmp/kde-ubuntu/simond/default/compile/hmm24/macros /media/104d991d-2062-40d7-89f6-ddde3cb5b781/home/ubuntu/Documents/2011-i/german-0.2.5/object/split/decc/german-speech-model-decc/macros-decc
d. Copy stats:
cp /tmp/kde-ubuntu/simond/default/compile/stats /media/104d991d-2062-40d7-89f6-ddde3cb5b781/home/ubuntu/Documents/2011-i/german-0.2.5/object/split/decc/german-speech-model-decc/stats-decc
e. Copy tiedlist:
cp /tmp/kde-ubuntu/simond/default/compile/tiedlist /media/104d991d-2062-40d7-89f6-ddde3cb5b781/home/ubuntu/Documents/2011-i/german-0.2.5/object/split/decc/german-speech-model-decc/tiedlist-decc
f. Get GPL license text:
wget http://script.blau.in/etc/GPL_License
g. Export the scenario (name: scenario-decc.xml) via simon > Manage scenarios > Export > Export to file
22. You now know how to compile a speech model. And you know how to get the files hmmdefs, tiedlist, macros, stats.
How to import a speech model
Friday, March 4th, 2011This article explains how to import the German speech model ‘debx’:
1. Remove old directories via Ubuntu terminal:
rm --recursive /home/ubuntu/.kde/share/apps/simon
rm --recursive /home/ubuntu/.kde/share/apps/simond
2. Stop simon. Stop simond.
3. Start simon.
4. Download the speech model, and extract it. On my computer, the extracted speech model is located at /home/ubuntu/Documents/german-speech-model-debx.
5. simon > Manage scenarios > Import > Import from file (location: /home/ubuntu/Documents/german-speech-model-debx/scenario-debx.xml)
6. Move the available scenario de-bx from the ‘Available’ column to the ‘Selected’ column. Press OK.
7. simon > Settings > Configure simon … > Model Settings > General > Static model. Set the paths to the static (base) model:
file:///home/ubuntu/Documents/german-speech-model-debx/hmmdefs-debx
file:///home/ubuntu/Documents/german-speech-model-debx/tiedlist-debx
file:///home/ubuntu/Documents/german-speech-model-debx/macros-debx
file:///home/ubuntu/Documents/german-speech-model-debx/stats-debx
Press Apply. Press OK.
8. Start ksimond. simon > Press the Connect button. simon is now activated. Let’s dictate a few words:
Gedankenreichtum Gedankenganges Gedankengangs Gebudeflche Geburtstagsparty Geburtstagspartys Geburtstagsfeier Geburtstagsfeiern Geburtshelfers Geburtenbeschrnkung Gebrauchtwagenabteilung Gebrauchtwagenhandel Gebrauchtwagenkaufs Gebrauchtwagenmarkt Gebrauchsfahrzeug Gebirgsbewohner Gebietsforderung Gebietskrperschaft Gattungsalternative
Almost all words were recognized correctly. But where are the German special characters (äöüß)? They are missing.
9. Press the Activated button to stop recognition.
10. You now know how to import a static speech model.
German speech model ‘debx’
Friday, March 4th, 2011Import the German speech model ‘debx’ into simon. The words are from section debx, and can be found at Voxforge, too.
German speech model ‘debs’
Friday, March 4th, 2011You can import the German speech model ‘debs’ into simon. The words are taken from section debs, and can be downloaded from Voxforge.
German speech model ‘debn’
Thursday, March 3rd, 2011Import the German speech model ‘debn’ into simon. The words are taken from section debn. You can get them from Voxforge, too.
