GiellaLT

GiellaLT provides an infrastructure for rule-based language technology aimed at minority and indigenous languages, and streamlines building anything from keyboards to speech technology. Read more about Why. See also How to get started, and our Privacy document.

View GiellaLT on GitHub

Page Content

Intro
Links

Languae recognition improvement

Intro

We now use text_cat. Todo:

Improve text_cat
Set up a testbench for evaluation
Test other language recogntion solutions.

Links

Sitemap

About GiellaLT
GiellaLT – Language technology for all
GiellaLT — Corpus Resources
GiellaLT — Dictionary Sources
GiellaLT — Keyboard Layouts
GiellaLT — Language games
GiellaLT — Language models
GiellaLT — Shared resources
GiellaLT — Speech technology resources
GiellaLT — Terminology Resources
Language resource maturity classification
Maintainers, Developers, Teams and Partners
Privacy
The table columns of the resource overviews
Where can I find documentation for what?
apps/ - GiellaLT apps:
- Gïelese documentation
- iOS App Development And Debugging
- Localising apps and webapps related to the GiellaLT infrastructure
- Updating Termwiki To Satni
- satni/ - satni.org:
courses/ - Courses and arrangements related to the giellalt infrastructure:
- nordisk/ - Kurs i installation och användning av språkteknologiska program för nordiska minoritetsspråk:
  - Praktiske opplysningar
  - Program for Kurs i installation och användning av språkteknologiska program för nordiska minoritetsspråk
dicts/ - The dictionary page:
- Arbeidsmåte,
- Compiling the Apertium-style webdicts
- Dictionary engines
- Dictionary manipulation
- Dictmeeting
- Documentation on satni.org
- Er ordboksmanuset mitt maskinlesbart?
- Estonian-Norwegian
- Find unlexicalised compounds in the dictionaries
- GEO er den geografiske automaten vår.
- Interactive Dictionary Compilation
- Kotusresources
- Lexicalisingnorwegian
- Lexicalisingsaami
- Lexicon files
- Literature on lexicography for minority languages
- Lulesamisk-norsk ordbok
- Målet med ordboksarbeidet
- Møte om ordbøker, terminologi og REST (el. GraphQL) 6.12.2017
- Nordsamisk-norsk ordbok
- North Saami dictionaries
- Nye eigenskapar i NDS
- Obsolete (old infra)
- Plains Cree dict
- Plan for å lage ei grammatisk ordbok
- Prinsippforordbokssnuing
- SJEKKLISTE for VD-ordboka (gammel infra):
- Sjekkliste før ordboka genereres og kompileres
- Spanish - North Saami dictionary
- Suggestions for NDS
- Testing
- Text for demo
- The first 10000 FO words
- Theoslobergentagger
- Top-2000 words from first run
- Webdict
- SkoltSaami2X/ - Skolt Sámi dictionary work:
  - Abbreviations
  - Adjectives
  - Adpositions
  - Inflections
  - Lexicon files
  - Numerals
  - Variants
- fad/ - Forvaltningsordbok - administrativ terminologi:
  - 110302
  - Bakgrunnsdokument
  - Dictionary for administrative language
  - FAD- og korpusmøte 13.11.2012
  - FAD- og korpusmøte 14.1.2013
  - FAD- og korpusmøte 19.12.2012
  - FAD- og korpusmøte 20.11.2012
  - FAD- og korpusmøte 20.11.2012
  - FAD- og korpusmøte 25.1.2013
  - FAD- og korpusmøte 4.1.2013
  - FAD-möte 14.5. 2013
  - FAD-möte 17.9.
  - FAD-möte 28.11.
  - FAD-møte 11.6.
  - FAD-Møte 12.10.2012
  - FAD-møte 14.8.2013
  - FAD-Møte 15.10.2012
  - FAD-møte 16.5.13
  - FAD-Møte 18.10.2012
  - FAD-møte 18.3.2013
  - FAD-Møte 19.9.2012
  - Fad-møte 2. april 2013
  - FAD-møte 20.6.
  - FAD-møte 21.5.13
  - FAD-møte 21.5.13
  - Fad-møte 22.8.2013
  - FAD-møte 25.6.
  - FAD-Møte 25.9.2012
  - FAD-Møte 31.10.2012
  - FAD-Møte 5.9.2012
  - FAD-Møte 6.11.2012
  - FAD-møte 6.3.2013
  - Fad-møte 8. april 2013
  - FAD-møte
  - Möte fad 18.4.2013
  - Møte, FAD, 30.8.2012
  - Møte, FAD-ordbok, Torsdag 16.8.2012 kl. 14
  - Ordparallellisering
  - Plan for administrative terminology (forvaltningsordbok)
  - Til stades:
  - Tirsdag 7.1.2014
- fad2/ - sma/smj FAD:
  - kmd-møte 10.3.15.
  - Mål
  - Møte om lule- og sørsamisk korpusinnsamling
  - Møte om lulesamisk term/ord-bok
  - Møte om sørsamisk kmd-prosjekt: 18.9.2017
  - Møte, nobsma-utviding KMD-prosjektet 27.11.2017
  - Møte, nobsma-utviding KMD-prosjektet 4.12.2017
  - Ord:
- jus/ - Juridisk terminologi - juridihkalaš terminologiija:
  - Møtereferat fra møte med UiT torsdag 19.januar 2012.
- nds/ - Neahttadigisánit:
  - API
  - Configuration
  - Designing resources
  - Flask software
  - Including Neahttadigisánit in a website
  - Management script (nds)
  - Moving languages between projects
  - NDS for Programmers
  - NDS localisations
  - Neahttadigisánit language pairs
  - Neahttadigisánit linguistic settings
  - Prod adn Dev Branch Notes
  - Projects in browsers statistics
  - Setting up NDS for development
  - Starting new language pairs
  - Testing scripts
  - Troubleshooting
  - Updating dictionaries
  - User logs
  - Wordform generation and analysis details (Paradigms)
  - referat/ - Meetings:
    - 160920
    - 240127
    - Møte for samordning av ordboksarbeid fram mot 9.2. 2013
- termwiki/ - Termwiki related docs:
  - Møte om Termwiki og ordbøker 7.6.2017
  - Neste steg i termwikiarbeidet
  - Risten2 & Termwiki
  - Terminologyprojects
  - Termwiki, Risten2 og termredigeringskurs
  - Termwiki-møte
  - Termwikimøte
games/ - Language games in the GiellaLT infrastructure:
- Word guesser game
infra/ - Technical Maintenance:
- Add new language with language data from an external repo
- Adding a new language to the Github infrastructure
- Anonymous svn
- Application infrastructure
- Arbeidsfordeling
- Background
- Build everything for voikko LO for release
- Build libvoikko as a dynamic universal binary with hfst support
- Build Libvoikko gramamrchecker for Macos 10.9
- Building and installing hunspell
- Building hfst-ospell for inclusion in Voikko
- Building HfstVoikko for LibreOffice on Tiger
- Building MacVoikko
- Building the Voikko extension for LibreOffice
- c. (For system administrators:) Setup on a multiuser server
- Compilation speed tests
- Compiling and using the analysers
- Compiling HFST
- Compiling Voikko with hfst
- Configuration of the web server
- Converting documents to Markdown
- cygwin
- Cyrillicsaamikeyboard
- Describing make
- Directory structure for the source code
- Docu-cgi-bin
- Documentation for gut
- Documentation infrastructure
- Don’t Panic!
- Evaluation of alternative build systems
- eXist installation (from svn)
- Fetch (check out) and set up the linguistic files
- Forrest-forbetringar
- Getting started with the GiellaLT infrastructure on a server
- Getting started with the GiellaLT infrastructure on Linux
- Getting started with the GiellaLT infrastructure on the Mac
- Getting started with the GiellaLT infrastructure on Windows
- Getting started with the GiellaLT infrastructure
- GFM Test Document
- Git commands
- Git usage notes
- Git-svn history cleanup
- Github maintenance
- gitignore
- Gtsvn
- Gut documentation
- Gut Usage Examples
- heading = large heading with text “heading”
- How to write documentation in Markdown
- Hva som skal være med
- Infrastructure upgrade and Github
- Install CLI tools using Homebrew
- Install CLI tools using MacPorts
- Install perl and its modules
- Install XCode
- Installing Foma
- Installing Linux on Windows
- Introduction
- Introduction
- Introduction
- jenkins on Fedora 23
- Kyrillisksamisktastatur
- Languae recognition improvement
- Lion installation notes
- Lucene
- MacPorts
- make remake
- Maverick upgrade notes
- Migrating data to Git LFS
- Migrating to git
- Multiple configurations and out of source building
- Newcompilerchecklist
- Notes on the bash shell
- Oldinfradocumentation
- Online analyser functionality
- Otherissuesonthemac
- Our svn repositories
- Oversikt
- Paradigm generation
- Plan for neste infra-oppgradering
- Private repos
- prop noun editing
- Quasicode For Keeping Track Of Transducers
- Remaking make using GNU autotools?
- REST and GraphQL API’s
- Run cgi programs locally for development
- Saamskaja
- satni.org
- Server programs and logs
- Servers and users
- Setting upa multiuser server
- Setup Emacs for GiellaLT
- Sitebuild problems
- Snow Leopard upgrade notes
- Startup guide for using svn
- Status and future of Xerox and other FST tools
- Steps in upgrading the bugzilla installation
- Test of the language technology documentation
- Testing in GiellaLT infra
- The interface for our web pages.
- Transducer infrastructure
- Unicode pain points
- Using XMLMind XM Editor (XXE) to edit dictionary files
- Version info for LibreOffice-Voikko and the libraries it depends on
- Versjonskontroll
- Virtual servers
- What is Nikola
- xserve
- Yosemiteupgradenotes
- infraremake/ - Infraremake:
  - Adding morphological test data
  - Build problems in $GTHOME/xtdoc/gtuit
  - Building And UsingA Weighted Speller
  - Compiling tools for abbreviation handling
  - Debugging in-source documentation
  - Doccomments — In-Source Documentation
  - Edmonton presentation
  - Example Of File Sizes With Optimisations
  - Fikse lus i egennavn (propernouns)
  - Filename correspondences in the 2015 transition
  - FST names before and after the 2014 revision
  - Getting started with the GiellaLT infrastructure
  - How To Configure And Optimise Spellers
  - How to control compounding in spellers
  - How to distribute and merge updates from the templates
  - How to enable tags in natural languages
  - How to use XML files as lexc sources
  - How to write your own shell scripts for testing
  - Howtosetuplanguagedocumentation
  - Ideas for testing
  - In-Source Documentation Specification
  - Infrastructure maintenance
  - List of defined automake conditionals
  - Lookup and composition - a direction primer
  - Naming and organisation of the makefiles
  - Newinfraoverview
  - Ny infra - opplæring del 2
  - Outline of major goals
  - Plan and status
  - Plan and status
  - Present building
  - Presentation of the Divvun and Giellatekno infrastructure
  - Samansetjingsrestriksjonar
  - Spellers in the GiellaLT infrastructure
  - Testscripts for use in the Giellalt infrastructure
  - The basic outline
  - The directory structure of the GiellaLT infrastructure
  - The Giellatekno & Divvun infrastructure
  - The new infra - a technical overview
  - Transducer names
- korp/ - Installation and update of Korp:
  - Bugs and improvements in Korp
  - How to update monolingual content
  - How to update Ordbild, the WordPicture
  - How to update parallel content
  - How to update the time graph
  - Korp installation
  - This document contains notes and ideas
- server/ - Introduction and overview:
  - Meänkieli adaptions in our infrastructure.
- system/ - System:
  - Accesstoprivategit
  - Adding or changing user and the password
  - Adding users on the Linux servers
  - CheckInstall
  - Configuration of svnserve
  - Divvun-server
  - Managing groups
  - Oversyn over prosessar og tenester vi har eller planlegg
  - Passwordless login by using ssh
  - Requirements
  - Sending svn e-mail manually
  - Services
  - Signing
  - SVN reorganisation
  - Testing tools for the sámi language technology project
- web/ - Web services and pages:
  - Presentasjon av paradigmer på internett
  - Reworkingthedivvunweb
  - Verktøy for forfattere
keyboards/ - Keyboard development:
- AndroidKeyboards
- Getting Started With Keyboard Development
- Installing keyboards
- Mobile keyboard app icons
- Plan For More Keyboards
- Tips For Computer Keyboard Development
- X11
lang/ - Language overview:
- Names and multilinguality
- The common Makefile and scripts
- common/ - Language models (transducers):
- smi/ - Topics:
ling/ - Tutorials:
- Agenda
- Brukjupyter
- Capitalisation
- Check and fix parallel corpus
- Conversion errors
- Converting .doc files
- Converting .txt files
- Converting html pages
- Converting PDF files
- Corpus analysis
- Corpus conversion process
- Corpus conversion
- Corpus DTD
- Corpus maintenance
- Corpus plan
- Corpus repositories
- Corpus Tools
- Docu-distesting
- Docu-lextesting
- Documenting the derivational tags & system
- File alignment plan
- Global flowchart
- How to get a better converting
- How to write disambiguation files
- Improve corpus alignment
- Improvement notes 2016
- Introduction to working on corpora at Giellalt
- Language Independent Tags In The Giella Infra
- Language recognition
- LexC-kurs (fst-kurs)
- Linguistic analysis with GiellaLT models
- Maintenance of parallel corpus
- Morpheme border markup
- Norwegian Whitepapers
- OCR reading
- Online-teavsttat sámegillii
- Ordbild
- Our disambiguator: vislcg3
- Parallel corpus conversion
- Preprocessor-usage
- Preprocessor
- Publications using SIKOR
- Regular expressions
- Setting up the environment
- Skip pages
- Tag standardisation
- Testing our grammatical tools
- Testing tools for the Sámi language technology project
- Testruns.paragstesting
- The <Correct!> corpora
- The Bible as a corpus
- The corpus page
- The list of foreign words
- Unicodenormalisation
- Utility regexes found in gt/common/src
- Wikipedia as a Corpus
- Wikipedia as a corpus
- XSL-files
- freiburg/ - Freiburg:
  - Elantiers
  - ELAN
  - Fst
  - Graid
  - Gtcorpus
  - Metadata
  - Tla
  - Toolbox
  - Transcription
- hist/ - Historical corpora:
  - Historical Sámi orthographies
  - Texts
- ordbild/ - Ordbild:
  - Comments for Tino 27.4.12:
  - Ordbild
- spoken/ - Overview:
  - LIA
localisation/ - Localisation:
- BCP-47
- Linux
- Translatingwindows
ml/ - Machine learning:
- Some background notes for work on machine learning
mt/ - The Machine Translation Page:
- Compiling Apertium files
- Configuring and updating MT Server
- courses/ - MT courses:
- infra/ - MT Infra:
- omegat/ - OmegaT resources and integration:
- smedeu/ - North Saami - German:
- smenob/ - North Saami - Norwegian:
- smesma/ - North Saami - South Saami:
- smesmj/ - North Saami - Lule Saami:
  - TRANSFER RULES
- smesmn/ - North Saami - Inari Saami:
proof/ - Documentation for spelling checkers:
- Dir structure
- Linguistic testing
- Meetings
- Obsolete documentation
- Proofing Tools Release Procedure
- Speller release procedure
- Spellers
- Stavekontrollar og fst-ar
- The Speller Error Models
- gramcheck/ - Grammar checker documentation:
- hyph/ - Hyphenators:
- installing/ - Installing spellers:
  - How to install speller and grammarchecker on Linux
  - Installing spellers on machines with a separate admin account
- nordplus/ - Nordplus stavekontroll – oversikt:
  - Norplus uke 10
  - Stavekontrolltesting og norplusprosjektet
- proofresearch/ - Research on proofing:
- spelling/ - Spellchecking in Common Apps and Systems:
tm/ - Translation memory:
- OmegaT
- Translation memories
tools/ - Development tools:
- A collection of examples
- About SubEthaEdit (SEE)
- An introduction to emacs
- Autshomato
- Balddalasteavsttat
- Bargovuohki
- Basic tools
- Basic usage
- Bugzilla introduction
- Cgevalscript
- Commands for grammar checker developers
- Conversion scripts
- CornerStone
- Crashcourse in Unix/Linux for new users
- Describing the project status
- Fiillatiippat
- FileMerge
- Foma
- Forrest layout
- Forrestdtd
- hfst-tokenise
- How to check out the GiellaLT files in Tromsø
- How to use emacs for our purposes
- How to use the morphological parsers
- How to use the vislcg3 parser
- Initial setup and checkout
- Initial setup
- Installing a North Saami speller for php-enchant
- Introduction
- Links, the table-friendly alternative to lynx
- Linux-enchant-libvoikko
- lookup2cg - script
- lookup
- Make or update spellers for Voikko+HFST
- Menus and tabs
- Mt.eng
- Mt.fin
- Mt.nob
- Mt.sme
- Nano
- Omegat.eng
- Omegat.fin
- Omegat.nob
- Omegat.sme
- Parallel texts
- Paralleltekster
- Paratekst
- PuTTY
- RCS, the single-user version control system for our .cgi files
- Rinnakkaistekstit
- Sentence alignment programs
- Sentence alignment
- Setting the TCA2 parameters
- Setting up the XMLMind XML Editor (XXE)
- Setup + requirements
- Special Emacs modes
- SSH
- Svnconv_remember
- Tca2_testruns.paragstesting
- tca2
- TRANSLATION CORPUS ALIGNER (TCA) 2
- Unix for linguists
- Unix for Lingvistar
- Unix ja Divvunkurssat
- Unix для лингвистов
- Useful commandoes for linguists
- Using SVN on a Windows machine: TortoiseSVN
- Usingthevexxmleditor
- UTF-8 conversion
- vislcg3
- WARNING!
- What is lynx?
- Windows
- ocr/ - OCR work at GiellaLT:
  - Bibal daihe Basse čala (1895)
  - Experimenting with Tesseract
  - Prosjekt: Friisrettskriving til 1979-rettskriving
  - Tesseract development
transcriptions/ - Transcription:
tts/ - Developing TTS:
- Building a voice
- Combining parts
- eSpeak NG Text-to-Speech
- Preparing text for TTS
- Recording text
- Splitting and aligning recording and text
- Text postprocessing
- Text processing / normalisation
- use/ - How to use the GiellaLT speech tools: