The Lab’s recent projects get featured on Concordia University’s website.
What do dyslexia in children and endangered languages have in common? Concordia graduate Gina Cook combines her expertise in linguistics and computer programming to tackle both challenges — and more…
iLanguage Lab is getting ready to launch FieldDB, a cloud based data entry app created for researchers at McGill, Concordia and University of California Santa Cruz. FieldDB is written in 100% Javascript and uses CouchDB, a NoSQL data store which scales to accomodate large amounts of unstructured data. CouchDB uses Map Reduce to efficiently search across data, a win-win for our clients. FieldDB uses fieldlinguistics and machine learning to automatically adapt to its user’s data. Most importantly, even though FieldDB is a WebApp that runs in your browser, FieldDB can run 100% offline. FieldDB will go into beta testing the first week of July. FieldDB will be officially launched in English and Spanish on August 1st 2012 in Patzun, Guatemala.
What is FieldDB?
FieldDB is a free, open source project developed collectively
by field linguists and software developers to make a modular,
user-friendly app which can be used to collect, search and share
your data.Who can I use FieldDB with?
- FieldDB is a Chrome app, which means it works on Windows,
Mac, Linux, Android, iPad, and also offline.- Multiple collaborators can add to the same corpus, and you
can encrypt any piece of data, keep it private within your corpus,
or make it public to share with the community and other
researchers.How can FieldDB save me time?
FieldDB uses machine learning and computational linguistics to adapt to your existing organization of the data which you import and predict how to gloss it. FieldDB already supports import and export of many common formats, including ELAN, Praat, Toolbox, FLEx, Filemaker Pro, LaTeX, xml, csv and more, but if you have another format you’d like to import or export, Contact Us.
What are the principles behind FieldDB?
We designed FieldDB from the ground up to be user-friendly, but also to conform to EMELD and DataOne best practices on formatting, archiving, open access, and security. For more information, see Section 6 of our white paper. We vow never to use your private data, you can find out more in our privacy policy.
Curious how it works? FieldDB is OpenSourced on GitHub
Thanks to all users who tried Spy Or Not, a gamified phonetics experiment designed to crowd source rating the imitation of UK British, Russian and South African accents. In two months since its launch the game has been installed by over 400 users on Android.
We were excited to see that localizing the app in Russian helped get us Russian speaking participants; 22% of all Android installs were in Russia as well as other Russian speaking satellites.
Since last April we have put our Speech Language Pathology interns Émie, Kim and Catherine hard to work learning Git, Gimp, ImageMagick, Sox, FFmpeg, Praat, Eclipse, XML and Java to bring the Bilingual Aphasia Test to touch tablets. The Bilingual Aphasia Test was created by Michel Paradis as well as other members of the Bilingual Aphasia community world wide. The Bilingual Aphasia Test is a normalized test containing 30 subsections to diagnose and treat bilingual/multilingual Aphasia patients. The BAT is available in more than 50 language pairs.
AndroidBAT is an interactive OpenSource application of the Bilingual Aphasia Test (BAT) stimulus book. AndroidBAT is a ‘virtual paper’ of the original BAT paper version. Unlike a computer application, the AndroidBAT is designed to simulate the flexibility of the original paper BAT, with the added benefit of allowing for a diversity of data collection integrated directly into the test. AndroidBAT allows recording of eyegaze and audio during a patient’s interview, without visible external camera or microphone, while providing more analyzable data (e.g., eye-gaze, audio, touch, etc.) than the paper format. Data can then be easily synced or shared with colleagues. AndroidBAT works on tablets and phones.
Want to use the BAT with your participants? You can download the Android Bilingual Aphasia Test on Google Play.
Our interns also braved Javascript and built the Bilingual Aphasia Test Scorer which allows clinicians to enter the data gathered during the BAT and get a patient profile on competencies such as phonology, morphology, syntax, lexicon, reading, writing, speaking, short term language memory and comprehension among others.
Twenty years ago we had two applications to score the BAT (Bilingual Aphasia Test), the PCBAT and the MacBAT. At the Academy of Aphasia annual meeting in October 2011 members of the Bilingual Aphasia Test community got together, and decided to make a web-based BAT scorer, that will run on any computer, on any mobile device, anywhere. Bilingual Aphasia Test Scorer on Chrome Store
Want to adapt the Bilingual Aphasia Test for your participants, or run similar experiments which collect eye-gaze or touch data? The Android Bilingual Aphasia Test is OpenSourced on GitHub
Twelve years after NAPhC1, where iLanguage Lab’s founder was one of the student organizers, the lab is proud to become the North American Phonology Conference’s first industry sponsor!
Mutsumi Oi, grad student at University of Ottawa won the sponsored door prize of a Praat script customized to her needs. The prize includes commented source code and a screencast explaining how the script works, and how to modify the script to tweek it for future use. Oi has until October 31 2012 to claim her prize, we are excited to work with her and will keep you posted if she decides to OpenSource her script.
Praat is an OpenSource phonetics software by Paul Boersma and David Weenink. It has been used by generations of linguists to automate phonetic analysis, export aligned transcriptions (textgrid), spectrograms, intonation contours and many other visualizations of sound. Praat can be run in a GUI, or on the command line. iLanguage Lab has integrated Praat into many of its Node.js experimentation server projects.
For some sample scripts: http://www.linguistics.ucla.edu/faciliti/facilities/acoustic/praat.html
In one week the Gamify project has gotten roughly 700 participants from around the world, including Kazakstan!
Of the new visitors (which we assume are coming to play for the first time) they are averaging 3.4 pages per visit, most are completing the experiment, which takes an average of 5 minutes to complete. We won’t know for a few weeks how many of the participants have usable data.
Spy or Not has seen visitors from around the world, most are viewing all three stages of the game.
Surprisingly, we had a few installs on the Android Market, many of whom also went through all three stages of the game.
Spy or Not installs on Androids also shows activity from around the world, most importantly where we need the most participants Russia, UK and South Africa.
Thanks everyone for playing and sharing our game, our goal is 500 participants from Russia, UK and South Africa. It takes only 5 minutes to spread the word, challenge your friends to beat your score!
So you think you would make a good spy?
iLanguage Lab is proud to present the release of Spy or Not, a gamified psycholinguistics experiment made in collaboration with the Accents Research Lab at Concordia University headed by Dr. Spinu.
It is commonly observed that some people are “Good with Accents.” Some people can easily imitate various accents of their native language, while others appear struggle with imitation. This research is dedicated to building free OpenSource phonetics scripts to extract the acoustic components of native speakers and “Good with Accents” speakers to transfer the technical details in a visualizable format to applied linguists on the ground who are working with accented (clinical and non-native) speakers.
In order to collect non-biased judgements from native speakers, a pilot study was designed and run by Dr. Spinu and her students. Images and supporting sound effects were created and the perceptual side of the pilot was disguised as the game “Spy or Not?” The game has since gathered over 8,000 data points by crowdsourcing the judgements to determine the degree (on an 11 point scale) of which participants were “Good with Accents.” This a novel approach to the coding problems that experimenters frequently encounter.
Participation in this project furthers research in phonetics and phonology in addition to experimental methodology in the age of the social web. Our hope is that our readers will Tweet their “Good with Accents” scores and help us get more participants, especially native speakers of Russian English accents, Sussex English accents and South African accents, accents we could never access at the scale we need in a lab setting. Visit the free online game, or play offline by downloading the game at the Chrome Store or on Google Play as a Android App.
If you have a spell checker, you want it to suggest a number of words that are close to the misspelt word. For humans, its easy for us to look at ‘teh’ and know that it is close to ‘the’, but how does the computer know that? A really simple Language Independent way to do it if you don’t have any gold standard data, is to assign costs to the various edits, substitution (2), deletion (1) and insertion (1), and picking the cheapest one.
The table below applies Levenshtein’s algorithm (basically, substitution costs 2) letter by letter. The total distance between the two words, 4 is in the top right corner, because it costs 2 to substitute ‘u’ for ‘i’ and 2 to substitute ‘t’ for ‘k’.
At the Lab, we put together an interactive javascript so that you can input whatever words you like and find out their edit distance. Just enter the words you want to compare!
And if you really like it, you can download it from github.
Click here to read more about edit distance.
Languages differ in how they overtly mark functions and their arguments, if they overtly mark at all… This month’s iLanguage game shows an example of the “Ergative-Absolutive” system, present in Hindi-Urdu, Walpiri, Inuktitut among others. In reality, its not as complicated as its name might indicate, in fact, we argue its quite logical.
In 9.10.a and 9.10.c we see that the Experiencer of travel is rather consistently is marked with -aq, as is Experiencer of greet in 9.10.b and 9.10.d. What might surprise you if you speak English, French or any other “Nominative-Accusative” language is that the -aq is consistently on the Experiencer, regardless of whether its the subject or the object.
In 9.10.e and 9.10.g we see that 1st person Experiencers appear on the verb, not as pronouns. This might sound familiar if you studied/speak Spanish.
9.10.f is particularly exciting since we don’t have enough data to say what is going on. We recommend stopping the next Yup’ik Eskimo speaker you run into and asking them to give you a verb that ends in a consonant, they might put -aq on the end when you give them a context to bring them to say he xe-ed yesterday…
Want to see more language data? Examples are taken from I-Language: An Introduction to Linguistics as Cognitive Science by Daniela Isac and Charles Reiss. http://linguistics.concordia.ca/I-language/
This month’s Guess the code shows an example of Functional Application which is an operation in compositional semantics. As you all know, in mathematics and in programming a function takes an input argument from some specified domain and yields an output value. Applying a function f to an argument x yields the value for that argument, which can be written as f(x). In beginner semantics, this same procedure is what happens when verb takes its object.
verb(object) ~ function(argument)
The only mystery in this example is the denotation double brackets which indicate that it is the denotation, not the orthographic word, which is being operated upon.
The code also gives us two trees to show that in English functional application applies to the right, and in Turkish it applies to the left.
In (41) we simplify things and pretend that “hug” is a function, which takes “Mary” as its object.” In reality, in most languages “hug” is a complex predicate, itself the return value of a Functional Application between a function (we call it “little v”) and its object, a root. Sound like Javascript anyone?
\begin{example}Typical example of Functional Appilcation (FA)\\
\label{typicalFA}
\begin{tabular}{ll}
\Tree[-1]{
\K{(a) English}\\
& VP_{}\Below{$_\textsc{fa}$}\B{dl}\B{dr} && =\denote{hug}(\denote{NP})\\
{V}_{>}\Below{\denote{hug}} && NP_e\Below{\denote{Mary}}\\
}&
\Tree[-1]{
\K{(b) Turkish} \\
& VP_{}\Below{$_\textsc{fa}$}\B{dl}\B{dr} && \hspace{-.2in} =(\denote{NP e})\denote{sardil-di}\\
NP_e\Below{\denote{Mary-e}} && {V}_{>}\Below{\denote{sarid-di}}\\
}
\end{tabular}
\end{example}
You can get the code here: https://github.com/iLanguage/ToolsForFieldLinguistics/blob/master/src/com/fieldlinguist/latex/latex.examples.tex