#language tech

This month’s linguistics jobs interview is with Brent Woo, a Language Engineer for a voice assistant. Brent has an MA from Eastern Michigan University and a PhD in Linguistics from the University of Washington. In this interview, Brent describes how his career provides him with work-life balance and how his linguistics training ties into his work developing technology we use every day, but might not realize requires language data to run smoothly. 

You can find Brent online on LinkedIn. 

What did you study at university? 

Linguistics and more linguistics! I entered undergrad not really knowing what I wanted to do. I took some finance and economics with the vague idea of becoming an accountant, but I didn’t have an interest in the material. I took some political science and cultural studies classes with some interest in “international affairs” but it still wasn’t quite right for me. I took an intro linguistics class and knew immediately this was it. This was the mix between analytical thinking and interesting human issues that I was looking for. I went on to finish BAs in Linguistics and Russian Studies, and then an MA and PhD in Linguistics. What is your job? 

Language Engineers work on the language designs for voice assistants. When you say something funny like “let’s raise the roof” and the voice assistant can interpret that to raise its volume to the maximum level, that was likely due to an LE language design. LE look into common variations on requests, understand the failures in human-machine communication, and adjust recognition coverage for them. We work on the scale of millions of utterances a day, so this process also has to be automated to some degree. How does your linguistics training help you in your job? 

I’m lucky to be considered a subject matter expert in language, and can usually make compelling arguments for utterance coverage based on language and human arguments alone. A common scenario is product or business stakeholders will request to have utterance X go to feature Y, and an LE will know from their expertise that no sane human customer would utter X to trigger Y. The LE has to present the case why that wouldn’t be a good design, and propose alternatives, and overall help the feature get better language coverage. This involves classic syntax for figuring out sentence structures and variations, and semantics for the annotation schema. Sometimes phonetics comes up when there are speech recognition conflicts or misfires! What was the transition from university to work like for you? 

Graduate school was great preparation for moving into this position. I already had experience running quantitative experiments and analyzing the results, managing different stakeholders (advisor, students, lab, external collaborators), and—most importantly—dealing with difficult customers. Two things were new to me at the beginning: having conversations at the business and product level, and the pace of work. I wasn’t familiar how to navigate conversations about project budget and headcount, but I had a good friend as a mentor who helped me out with these early on. The pace of work differs from grad school: work comes and goes with the product release cycles, there are periods of crunch time and periods of relative calm and catch-up. Do you have any advice you wish someone had given to you about linguistics/careers/university? If you’re interested in exploring a non-academic path, talk to linguists in those paths, watch the panels and webinars we’ve run on these topics, ask for mock interviews, and read Superlinguo! You’ll get a sense of what background you need and what projects you should be working on. 

Any other thoughts or comments?

The most important thing for me when finding a job after graduate school was to find a work-life balance. While I’m happy and lucky to have an interesting job related to language, I was really thrilled to discover that it was very flexible and I can live a rich life outside of work responsibilities. There were other jobs that may have been a better fit, but were more demanding on my lifestyle or diplomatic capacity (for example, managing people). I have many colleagues at this and other companies who have taken on more intense positions and end up burning out or quitting due to the stress. I find the work-life balance in this position almost ideal and I am very happy.

Related interviews:

Interview with a Natural Language Annotation Lead

Interview with a Software Engineer

Interview with a Data Scientist

Interview with a User Experience (UX) Researcher

Interview with a Computational Linguist

Recent interviews:

Interview with a Data Manager & Digital Archivist

Interview with a Natural Language Annotation Lead

Interview with an EMLS/Linguistics instructor & mother of four

Interview with a Performing Artiste and Freelance Editor

Interview with a Hawaiian and Tahitian language Instructor, Translator & Radio Host

Resources:

The full Linguist Jobs Interview List

The Linguist Jobs tag for the most recent interviews

The Linguistics Jobs slide deck (overview, resources and activities)

If you were to walk down a street in New York City, you might encounter hundreds of fonts on a single block: everything from street signs to shop names take on distinct personalities. Urdu and its nastaʿlīq typeface, lagging in the digital space, are much more difficult to iterate on with existing technology. Where there are hundreds of thousands of Latin digital fonts — fonts so iconic that we know them by name, like Helvetica or Times New Roman — digital Urdu fonts are limited to a handful at most. The nastaʿlīq typography that you might see on the streets of Pakistan is most likely to be hand painted.

“We went on Google and typed “what are fonts, how do we develop them,” Zeeshan recalled. “We purchased software but sat on it for a long time because we didn’t know how to use it.” Produced in collaboration with a government technology board over the course of a decade, Mehr Nastaliq uses 500 characters, all handwritten by Nasrullah — a tiny fraction of the 20,000 glyphs Jamil so painstakingly wrote in the 1980s. The Mehrs are particularly proud of how light their font is: at 60 KB, it doesn’t slow down websites and renders quickly. You can elongate letters and add diacritics.

Their experience of developing Mehr Nastaliq drove home an important lesson: the need for close collaboration between the calligrapher and developer. “It is imperative that the calligrapher and developer understand each other,” emphasised Zeeshan. “Abbu understands programming too now, so he can offer a different solution when confronted with tech limitations.” Working with his parent comes with its own challenges, of course. “Sometimes, Zeeshan would say, This letter isn’t looking nice,” Nasrullah said with a chuckle. “And I’d get huffy and say, Are you the calligrapher?”

Unicode, developed in the early 1990s, is now a global standard for representing characters across language systems in computing code, which means if you write Urdu — in naskh or nastaʿlīq — on one computer, it won’t appear as a string of garbled symbols in another. But technological advances don’t automatically bode well for the digitization of non-Latin type languages. Nemeth notes that the proliferation of easy-to-use design software doesn’t eliminate the need for script-specific expertise. “Designers who are not willing or able to invest the years of learning and research necessary to master a foreign script,” he says, “are led to believe that their tools and some superficial ‘borrowing’ of design elements are sufficient for successful design.”

An article in Slate about language revitalization in the digital era. Excerpt: 

At a recent conference called Breath of Life 2.0, held at Miami University of Ohio, participants explored the possibilities and pitfalls of archival databases. For the people involved in language revitalization work, whether they’re awakening a dormant language (one that no longer has fluent speakers) or trying to prevent an existing language from losing all its speakers, one of the major obstacles is a lack of digital resources.

Jerome Viles experienced that firsthand. An enrolled member of the Confederated Tribes of Siletz Indians and part of the Southwest Oregon Dene Languages Project team, Viles helps organize archival materials produced by linguists and community members over the past 100 years. First, his community had to findthe documents, which were scattered in institutions across the country. But the other problem was organizing and reconciling them. For instance, some records of language may have been written by French Jesuits who came to North America to convert Native American people. But even if three Jesuits worked with the same tribe, they could all have different spelling systems. Then there are audio recordings from the 1800s and 1900s, plus ethnographic material collected by anthropologists.

What Viles’ group and many others needed was a way to collect and compare thousands of words across multiple generations and several dialects. And that’s where the Indigenous Language Digital Archive came in. ILDA is a database built by engineers at Miami University in collaboration with the Miami Tribe of Oklahoma (which built an earlier form of the database specifically for its own language revitalization work). At the Breath of Life 2.0 conference, participants learned how to navigate the software and move their documents into one big digital space. By the end of the five days, a few surprising discoveries had emerged. Community members from the Northern Paiute, working on the Numu language, were tickled to find seven different ways to say “our husband.”

“I was getting a kick out of that,” said Nicholas Cortez. “I was like, where’s this ‘our’ coming from?”

Emily Bender on the coining of the Bender Rule and why it’s important to name the languages we make technology for. Excerpt: 

Though English and Mandarin Chinese are widely spoken, both as first and second languages, clearly a world in which advanced language technology exists only for these two languages is undesirable. The promise of language technology includes pro-social applications spanning a wide range from biomedical applications (e.g. matching patients to research studies or automatically flagging patients for time-sensitive tests based on physician notes), through machine translation of documents available on the web, to interactive tutors for language learning and other learning scenarios and more. These benefits should be available to all. 

Furthermore, the existence of even the most basic language technology (keyboards or input systems supporting the writing system, spell checkers, web search) builds up the value of a language which can be an important factor in self-esteem and educational outcomes for speakers of minoritized languages and can contribute to the maintenance of languages under threat of displacement by local majority languages (see e.g. Bamgbose 2011).

And yet, the field of NLP is caught in a negative feedback loop that hinders the expansion of the languages we work on. Work on languages other than English is often considered “language specific” and thus reviewed as less important than equivalent work on English. Reviewers for NLP conferences often mistake the state of the art on a given task as the state of the art for English on that task, and if a paper doesn’t compare to that then they can’t tell whether it’s “worthy”.

I believe that a key underlying factor here is the misconception that English is a sufficiently representative language and that therefore work on (just) English isn’t also language-specific. This misconception is abetted by the habit of failing to name the language studied when it is English.

A paper from the Workshop on the Use of Computational Methods in the Study of Endangered Languages (ComputEL-3) about making better tech tools for linguistic annotation, featuring many ideas that I wish I’d had in grad school! 

Abstract: 

To reduce the annotation burden placed on linguistic fieldworkers, freeing up time for deeper linguistic analysis and descriptive work, the language documentation community has been working with machine learning researchers to investigate what role technology can play, with promising early results. This paper describes a number of potential follow-up technical projects that we believe would be worthwhile and straightforward to do. We provide examples of the annotation tasks for computer scientists; descriptions of the technological challenges involved and the estimated level of complexity; and pointers to relevant literature. We hope providing a clear overview of what the needs are and what annotation challenges exist will help facilitate the dialogue and collaboration between computer scientists and fieldwork linguists.

This month’s interview highlights the increasingly central role of human language in tech, and the important role that linguists can play in the current tech landscape. It’s also been a delight for me to interview Sasha Wilmoth, who I last caught up with when she was taking one of her final undergraduate classes at Melbourne Uni. This interview is also something  slightly different, because Sasha is one of a growing number of ‘hybrid’ researchers - working across both the university and private sector. I’ll leave it to her to explain more!

What did you study at university?

I did a Bachelor of Arts majoring in Linguistics, and studied Mandarin and Swedish. I also squeezed in as many literature subjects as I could. Within linguistics, I became interested in the morphology and syntax of Australian Aboriginal languages – my honours thesis was on Murrinhpatha, which is a polysynthetic language spoken in the Northern Territory.

What is your job?

That’s a good question! I think most of my family and friends still don’t really understand what I do for a living.

My time has been split between two jobs this year: I’m a Linguistic Project Manager at a company called Appen, and a Research Assistant within the ARC Centre of Excellence for the Dynamics of Language (CoEDL). I’ll just focus on my job at Appen, because it’s a bit further away from academia.

Basically, Appen provides all sorts of data for AI and machine learning, including for speech and language technology. In the Linguistic Services department, we make the specialised annotated data that’s used to train these systems, in whatever language our clients need (we’re up to something like 180 languages and dialects by now). This includes pronunciation lexicons, prosodic annotation, part-of-speech tagging, tokenisation, proofing tools, guidelines, and so on. There are also lots of linguists working in the Data Collection, Transcription, and Translation departments. My job encompasses the technical and linguistic bits, as well as the more typical project management things like planning, budgeting, and delegating tasks.

For example, for a pronunciation lexicon project, which is what I’ve largely been working on, I start by researching the grammar, phonology, orthography and dialects of the language. Based on this research, I figure out what our approach to phonemic transcription will be and come up with some technical tools to predict the pronunciation of the words based on the orthography. I manage a team of linguists around the world who are native speakers of that language, and they check that the pronunciations are correct. Then I do a bunch of quality checks according to what we know about the language and prepare the data for our clients. Recently, I’ve been working on Serbo-Croatian, Persian, and Mandarin projects – as you can imagine, each language presents completely different technical and linguistic challenges.

Appen is an industry partner with CoEDL, so another cool thing I get to do is collaborate with academic linguists and see if we can apply our own processes and problem-solving skills to their research projects.

How does your linguistics training help you in your job?

I’m happy to say that I apply my linguistics education at work every day: I solve linguistic puzzles and read grammars and write phonological rules and use ToBI (a way of transcribing intonation) and the International Phonetic Alphabet. I never managed to become a very fluent Mandarin speaker, but I use that knowledge for our Mandarin and Cantonese projects. I even do a bit of Swedish proof-reading here and there. I didn’t imagine this was possible when I chose this rather impractical combination of subjects.

Do you have any advice you wish someone had given to you about linguistics/careers/university?

I know some previous interviewees have said they wish they had been more aware of the (lack of) job prospects for linguists. I’m stubborn though, and I was determined to study what I was most interested in without taking into account career prospects. I’m really glad I made that choice (and had the privilege to do so), because there are actually more opportunities outside of academia than I thought. The future for linguists in industry is bright, as we’ll be using speech more and more to interface with our phones, cars, toasters, personal robot servants, etc.

Most linguistics students don’t end up working as linguists, so it’s wise to keep your options open. But if you’re motivated and can’t think of anything else you’d rather be doing, there are jobs out there – I wish I had known about them when I was at uni. It’s really important to seek out and take advantage of opportunities for students like summer research programs, internships, or volunteering with organisations like RNLD here in Melbourne. Get to know your lecturers and tutors and ask them what you can get involved in.

Any other thoughts or comments?

Just a note on working in tech: I don’t have a computer science background, and I’ve never been a big computer nerd. I’ve learned everything on the job, and it turns out I quite like computers after all. If you have the type of pattern-finding brain that linguists tend to have, chances are you might enjoy the technical side of things when it’s applied to something interesting. Developing some technical skills as a student can help you in many career paths, but it’s not an absolute necessity. If you’re looking for a place to start: regular expressions, bash, and Python are all very useful and relatively easy to learn.

Previously:

Interview with a Data Scientist

Interview with a Librarian

Interview with a Text Analyst

Interview with a User Experience (UX) Researcher

Interview with a Study Abroad Facilitator

Interview with The Career Linguist

Interview with a local radio Digital Managing Editor