Shester Gueuwou

Shester Gueuwou, Rose-Mary Owusuaa Mensah Gyening

On December 7, 2020, Ghanaians participated in the polls to determine their president for the next four years. To gain insights from this presidential election, we conducted stance analysis (which is not always equivalent to sentiment analysis) to understand how Twitter, a popular social media platform, reflected the opinions of its users regarding the two main presidential candidates. We collected a total of 99,356 tweets using the Twitter API (Tweepy) and manually annotated 3,090 tweets into three classes: Against, Neutral, and Support. We then performed preprocessing on the tweets. The resulting dataset was evaluated using two lexicon-based approaches, VADER and TextBlob, as well as five supervised machine learning-based approaches: Support Vector Machine (SVM), Logistic Regression (LR), Multinomial Naïve Bayes (MNB), Stochastic Gradient Descent (SGD), and Random Forest (RF), based on metrics such as accuracy, precision, recall, and F1-score. The best performance was achieved by Logistic Regression with an accuracy of 71.13%. We utilized Logistic Regression to classify all the extracted tweets and subsequently conducted an analysis and discussion of the results. For access to our data and code, please visit: https://github.com/ShesterG/Stance-Detection-Ghana-2020-Elections.git

Shester Gueuwou, Sophie Siake, Colin Leong et al.

Advancements in sign language processing have been hindered by a lack of sufficient data, impeding progress in recognition, translation, and production tasks. The absence of comprehensive sign language datasets across the world's sign languages has widened the gap in this field, resulting in a few sign languages being studied more than others, making this research area extremely skewed mostly towards sign languages from high-income countries. In this work we introduce a new large and highly multilingual dataset for sign language translation: JWSign. The dataset consists of 2,530 hours of Bible translations in 98 sign languages, featuring more than 1,500 individual signers. On this dataset, we report neural machine translation experiments. Apart from bilingual baseline systems, we also train multilingual systems, including some that take into account the typological relatedness of signed or spoken languages. Our experiments highlight that multilingual systems are superior to bilingual baselines, and that in higher-resource scenarios, clustering language pairs that are related improves translation quality.

Serpil Karabüklü, Kanishka Misra, Shester Gueuwou et al.

Models of sign language have historically lagged behind those for spoken language (text and speech). Recent work has greatly improved their performance on tasks like sign language translation and isolated sign recognition. However, it remains unclear to what extent existing models capture various linguistic phenomena of sign language, and how well they use cues from the multiple articulators used in sign language (hands, upper body, face). We introduce a new benchmark dataset for American Sign Language, ASL Minimal Translation Pairs (ASL-MTP), divided into multiple types of sign language phenomena and corresponding minimal pairs of translations, for performing such linguistic analyses. As a case study, we use ASL-MTP to analyze a state-of-the-art ASL-to-English translation model. We conduct a targeted analysis of the model by ablating various input cues during training and inference and evaluating on the phenomena in ASL-MTP. Our results show that, while the model performs above chance level on most of the phenomena, it relies strongly on manual cues while often missing crucial non-manual cues.

Shester Gueuwou, Xiaodan Du, Greg Shakhnarovich et al.

Sign language processing has traditionally relied on task-specific models, limiting the potential for transfer learning across tasks. Pre-training methods for sign language have typically focused on either supervised pre-training, which cannot take advantage of unlabeled data, or context-independent (frame or video segment) representations, which ignore the effects of relationships across time in sign language. We introduce SHuBERT (Sign Hidden-Unit BERT), a self-supervised contextual representation model learned from approximately 1,000 hours of American Sign Language video. SHuBERT adapts masked token prediction objectives to multi-stream visual sign language input, learning to predict multiple targets corresponding to clustered hand, face, and body pose streams. SHuBERT achieves state-of-the-art performance across multiple tasks including sign language translation, isolated sign language recognition, and fingerspelling detection.

Shester Gueuwou, Xiaodan Du, Greg Shakhnarovich et al.

A persistent challenge in sign language video processing, including the task of sign to written language translation, is how we learn representations of sign language in an effective and efficient way that preserves the important attributes of these languages, while remaining invariant to irrelevant visual differences. Informed by the nature and linguistics of signed languages, our proposed method focuses on just the most relevant parts in a signing video: the face, hands and body pose of the signer. However, instead of fully relying on pose estimation from off-the-shelf pose tracking models, which have inconsistent performance for hands and faces, we propose to learn a representation of the complex handshapes and facial expressions of sign languages in a self-supervised fashion. Our approach is based on learning from individual frames (rather than video sequences) and is therefore much more efficient than prior work on sign language pre-training. Compared to a recent model that established a new state of the art in sign language translation on the How2Sign dataset, our approach yields similar translation performance, using less than 3\% of the compute.