BAMU · Third Year B.E./B.Tech Artificial Intelligence and Data Science · NEP 2020
Speech and Language Processing
PECT-2 · Elective III · Semester VI · 3 credits · CIA 40 + SEE 60
One of the Elective III options.
Module 1: Introduction; Word Level Analysis
15 hoursOrigins and challenges of NLP, grammar-based and statistical language models, regular expressions and finite-state automata, English morphology and transducers, tokenisation, spelling correction and minimum edit distance; n-grams, smoothing, interpolation and backoff, word classes, rule-based, stochastic and transformation-based POS tagging, HMM and maximum-entropy models.
Examined in the 60-mark end-semester paper
- Origins and challenges of NLP
- Grammar-based and statistical language models
- Regular expressions and finite-state automata
- English morphology and finite-state transducers
- Tokenisation, spelling correction and minimum edit distance
- N-grams, smoothing, interpolation and backoff
- Part-of-speech tagging: rule-based, stochastic and transformation-based
- HMM and maximum entropy taggers
Module 2: Syntactic Analysis; Semantics and Pragmatics
15 hoursContext-free grammars and English treebanks, normal forms, dependency grammar, dynamic-programming and shallow parsing, probabilistic CFGs and CYK, feature structures and unification; first-order and description logics, syntax-driven semantic analysis, word senses, thematic roles and selectional restrictions, word sense disambiguation and word similarity.
Examined in the 60-mark end-semester paper
- Context-free grammars and treebanks
- Dependency grammar
- Dynamic programming parsing and shallow parsing
- Probabilistic CFGs and probabilistic CYK
- Feature structures and unification
- First-order logic and description logics for meaning
- Syntax-driven semantic analysis and thematic roles
- Word sense disambiguation and word similarity
Module 3: Basic Concepts of Speech Processing; Speech Analysis and Modelling
15 hoursArticulatory and acoustic phonetics; the short-time Fourier transform, filter banks and LPC; speech distortion measures, log-spectral and cepstral distances, LPC, PLP and MFCC features; time alignment and dynamic time warping; HMMs for speech — evaluation, Viterbi search and Baum–Welch re-estimation.
Examined in the 60-mark end-semester paper
- Articulatory and acoustic phonetics
- Short-time Fourier transform, filter banks and LPC
- Speech distortion measures and cepstral distances
- LPC, PLP and MFCC features
- Dynamic time warping
- HMMs for speech: evaluation and Viterbi decoding
- Baum-Welch re-estimation
Module contents are summarised from the official BAMU syllabus . AiCanCode.org is not affiliated with Dr. Babasaheb Ambedkar Marathwada University (BAMU).