Learn through hands-on exercises how to balance theoretical and practical aspects of natural language processing. This course covers both text and speech data.

Syllabus

Introduction

Fundamentals of natural language processing
NLP course strategy

1. Introduction to Natural Language Processing (NLP)

What is natural language processing (NLP)?
What are sequences?
Applications of natural language processing in text data
Applications of natural language processing in speech data
Historical evolution of NLP tasks and techniques
How computers understand sequences in NLP

2. Natural Language Processing for Text Techniques

Text preprocessing
Text preprocessing using NLTK
Text representation
Text representation: One-hot encoding
One-hot encoding using scikit-learn
Text representation: N-grams
N-grams representation using NLTK
Text representation: Bag-of-words (BoW)
Bag-of-words representation using scikit-learn
Text representation: Term frequency-inverse document frequency (TF-IDF)
TF-IDF representation using scikit-learn
Text representation: Word embeddings
Word2vec embedding using Gensim
Embedding with pretrained spaCy model
Sentence embedding using the Sentence Transformers library
Text representation: Pre-trained language models (PLMs)
Pre-trained language models using Transformers

3. Natural Language Processing for Speech Techniques

Speech representation: Mel-frequency cepstral coefficients
Mel-frequency cepstral coefficients (MFCCs) using librosa
Speech representation: Linear predictive cepstral coefficients (LPCCs)
Linear predictive coding (LPC) using librosa
Speech representation: Gammatone filterbank features
Gammatone filterbank features using librosa
Speech representation: Spectrograms
Spectrograms using fast Fourier transform (FFT) in librosa
Speech representation: Speech embeddings
Speech embeddings using wav2vec in Transformers

4. Applied Natural Language Processing: Algorithms and Tasks

Algorithms for natural language processing tasks
Types of algorithms in natural language processing
Rule-based: Regular expressions
Regular expression tasks using the re library
Rule-based: Rule-based parsing
Parsing sentences into syntactic structures using context-free grammars (CFG) in NLTK
Part-of-speech (POS) tagging using spaCy
Statistical: Hidden Markov models (HMMs)
Hidden Markov models (HMMs) for POS tagging in NLTK
Statistical: Conditional random fields (CRFs)
Statistical: Naive Bayes classifiers
Machine learning: Support vector machines (SVMs)
Classify text data using SVM
Machine learning: Decision trees
Classify the speech commands dataset using decision trees
Machine learning: K-means clustering
K-means clustering for the movie reviews dataset
Deep learning: Recurrent neural networks (RNNs)
Text generation using recurrent neural networks (RNNs)
Deep learning: Transformers
Transfer learning in natural language processing (NLP)
Speech-to-text (STT) using wav2vec in the Transformers library
Text-to-speech (TTS) using Tacotron and WaveGlow