Ricardo portrait

Computational Linguistics Robust NLP & Speech Processing Researcher

About me

I’m Tatiana, a Language Sciences MSc student with a deep passion for computational linguistics and language technology. While my dissertation focuses on speech emotion recognition (SER), my broader interest lies in exploring the challenges posed by noisy data in language technology. I’m particularly curious about how large language models (LLMs) perform in noisy environments and how they can be adapted for low-resource languages. As humans, we often communicate imperfectly, and I believe that developing robust systems to handle this linguistic noise will help advance language technologies.

Research Interests

My research interests are centred around natural language processing (NLP), machine learning, and computational linguistics, with a strong focus on working with noisy data. I am particularly interested in how LLMs behave under noisy conditions and how they can be used for classification tasks in low-resource languages. My proposed research explores the potential of KB-BERT for handling noisy Swedish text, investigating techniques such as adversarial training to improve performance in environments with typographical errors, informal language, and phonetic misspellings. I am also keen to explore how NLP solutions can be scaled and adapted for low-resource languages, aiming to create systems that can handle both linguistic noise and limited resources effectively. Ultimately, I want to contribute to building language technologies that are more resilient and adaptable to the imperfections of real-world communication.

My skills

Technical Skills

  • Programming Languages:Python, R
  • Machine Learning:TensorFlow, PyTorch, Scikit-learn, Keras
  • Natural Language Processing (NLP):SpaCy, NLTK, re
  • Data Analysis:Pandas, NumPy, Matplotlib, Jamovi
  • Version Control:Git, GitHub

Language skills

English
Native (C2)
100%
Russian
Native (C2)
100%
Spanish
Native (C2)
100%
Catalan
Upper Intermidiate (B2)
75%
French
Intermidiate (B1)
60%
Swedish
Elementary (A2)
40%

My experience

Certifications

Linear Algebra Course

Codecademy, Issued Jan 2025

Text Preprocessing Course

Codecademy, Issued Nov 2024

Machine Learning: Advanced Learning Algorithms

DeepLearning.AI, Issued Oct 2024

Skills: TensorFlow, Artificial Neural Networks, Model Development

Learn Python 3 Course

Codecademy, Issued Sep 2024

Credential ID: 66E828F09D

Skills: Python, Computer Science

Supervised Machine Learning: Regression and Classification

DeepLearning.AI, Stanford University, Issued Sep 2024

Credential ID: DJ79GXQYEYY6

Skills: Python, Machine Learning

Learn the Command Line Course

Codecademy, Issued Aug 2024

180 Hour Level 5 TEFL Certificate Course (Highfield)

Premier TEFL, Issued Jan 2023

Contact me

Please contact me directly through this form.