Julia Hirschberg

It's a very low workload class, but not sure how much I learned. It has no exams, all grades are from reading responses and homeworks. The homeworks don't require any knowledge from class, just basic Python and ML, so I wish the homeworks related a bit more to speech tehcniques and what is covered in class. The homeworks are somewhat tedious as there is code but they also make you write written and repetitive responses for each section (and they take points off if you don't write a lot). Sometimes feels like TAs want us to just do busy work rather than actual interesting work. But it isn't a high workload class, I just didn't love it.

Awesome professor who in any other context may be mistaken for a cat lady. She puts interests in every one of her students (admittedly easier than others because of a class size of 15 or so), is easily approachable, and keeps the long (2 hr) discussions as fun as they could possibly be. Also one of the best academics in her field... if you're a CS major looking to do coursework in NLP, you should absolutely take at least one class from her!

This is not the class I thought I signed up for, but now that I'm done with it I think it was worth taking. It's insufficiently clear from the course directory that NLP (at least with Prof. Hirschberg) is entirely a survey course: you learn very little that you can sit down and implement, and a lot about broad categories of problems in computational linguistics and the general approaches that has been used to solve them. There are a handful of algorithms I'll hang on to---two months after the midterm I'm fairly confident I could write a Brill tagger or a CYK parser---but realistically, I'll never need to, and most of the material is not that specific. This class is not so much about learning to do NLP as it is about learning to understand what's already been done. Prof. Hirschberg herself is very sweet---helpful and accessible outside of class. She also has a very soothing voice. This turned out to be a bit of a problem: I came every day, but to my great shame would start to doze off around the 45-minute mark almost half the time. I would love for her to read me bedtime stories, but the bottomless slide decks and repetitive nature of the material mean that the lectures are unstimulating to the point of peril. If you're at all interested in NLP I'd certainly recommend that you take this class; if you're just trying to fill the AI track requirements you'll learn more impressive…

I took this class in Fall 2022. If you want a really simple, interesting class then I'd say go for it. Julia is a very sweet professor (I wish she offered more classes), but I'd be lying if I said I attended more than 3 lectures in person. The only lecture necessary for doing (pretty much all) the homeworks was the one in which she gives a run-down on Praat, the software used for the course. That being said, the lectures are pretty interesting. The homeworks are super easy if you've done any kind of work using Python and know some supervised learning techniques (even if you don't I feel like this should be pretty easy to learn). You can get away with starting homeworks 2 days before submission, and you definitely don't need more than 4 days for a homework. It may take a while to read through the documentation of the software (it was a little difficult to understand how to use Praat with Python) and get an outline of what to do for the homework, but once you have that, it's pretty straightforward. You also might have some trouble figuring out how to handle huge amounts of data, and it might be time-consuming, but I really don't think it's a tough course at all. The only time I've had a major problem is when I wasn't able to download some software used for the course and none of the TAs really seemed to have a solution since I was on a Windows OS (I think it's pretty simple for…

Weekly paper readings (varies in length), 1 Final Project (Proposal, Interim Report, Final Report and Presentation)

Man, I must not have been taking the same class as the other two reviewers. NLP is an interesting topic but somehow it ended up getting reduced to baby talk, with endless probing of the uninterested students for examples to incredibly simple questions. She's a nice person and all, but this felt like a big waste of time.

Awesome. Her and Professor Gross are my favorite professors. Her classroom delivery is fantastic. Slides were lacking in content but she more than made up for it. If you attend class you'll be perfectly fine. Notes don't really convey audible concepts well so it helps to hear them in person. I think you'd get pretty lost if you didn't attend regularly. At least skim through the assigned readings, I found actually reading them to be unnecessary. Just getting the gist of it or reading the abstract is usually ok. The lectures are the theory, the homework is the implementation, which makes it sometimes necessary to visit the TA to find out how to do it.

An excellent professor, not a bad class. Professor Hirschberg is engaging, friendly, and interesting, incredibly well prepared for class, and impressively knowledgeable about her field. You should expect the class to tend towards your humanities side, if you are used to the standard comp. sci class. Exams consist of essays rather than calculations, and at least half the class is devoted to the study of natural language itself, often seperate from computation. There are very few algorithms to learn, material is approached on a "higher level" that avoids technical knowledge. Going to class is not strictly required, since there are beautiful lecture notes for each class, as well as chapters and sections in the book posted for each lecture. Missing class however would be a mistake, as she is an engaging lecturer.

Reading responses. Pros and cons for each chapter or each reading. This is easy but annoying as sometimes the actual chapter we were referencing was shorter than the 200 word responses we'd have to write (for EACH chapter). 3 homeworks throughout the semester, and they won't even have graded Hw1 until 1 week before the end of the semester. No exams

Julia is, in a word, amazing. She is a really sweet, nice lady who is also extremely knowledgeable about her field and seems to really LOVE teaching. Her melodic speaking voice makes her sound like she's talking down to the class, but don't be fooled; once you get used to her you realize that she truly respects her students. The classes are a mix of lecture and discussion, but tend toward discussion most of the time. Julia really encourages the class to get involved (there is a class participation grade), and usually there are many great insights into the material. She will learn your name, and you are always welcome to chat with her after class or in office hours. If you are at all interested in NLP, or you think you might be, do not miss this class.