ID:
511850
Duration (hours):
36
CFU:
6
SSD:
GLOTTOLOGIA E LINGUISTICA
Year:
2025
Overview
Date/time interval
Primo Semestre (22/09/2025 - 19/12/2025)
Syllabus
Course Objectives
The course will introduce students to the use of the most modern techniques for eliciting, recording and analyzing speech (read and spontaneous). At the end of the course the students will have achieved a series of theoretical and practical knowledge useful to better comprehend and analyse the spoken dimensions of the languages or of the language varieties from different perspectives (computational, sociolinguistic, ethical, etc.). In addition, students will know how to orient themselves in the specialist literature and they will be able to independently conduct and present theoretical and/or experimental research in the field of phonetics, speech processing, automatic speech recognition (ASR), and speech synthesis / text-to-speech (TTS).
Course Prerequisites
The course is open to all the students interested in examining phonetics and speech processing in depth. To better face this course, basic knowledge of phonetics and phonology are welcome (knowledge acquired, for instance, in a course of general linguistics), in particular the International Phonetic Alphabet (IPA). To brush up on the basic notions of phonetics, the following textbook is recommended: - Nespor, M. & Bafile, L. (2008). I suoni del linguaggio. Bologna: il Mulino. [chapters: I, II, III]. Mandatory knowledge of Python programming or NLP models and systems development is not required. A rudimentary understanding of the basic concepts of computational linguistics is helpful but not necessary.
Teaching Methods
The course features both face-to-face lectures and practical exercises. The lectures will be supported by Power Point presentations that will be projected in the classroom and they will be accompanied by supporting multimedia material. The course will also consist of several laboratories in which students guided by the instructor will work on hands-on activities and assignments, often in small groups. The practical exercises will give students the opportunity to try their hand with experimental phonetics and speech processing (e.g., recordings of spoken and spontaneous speech, segmentation of audio material, extraction and annotation of features, acoustic analysis, speech classification, ASR, TTS, etc.). The practical exercises will take place in collaboration with the Experimental Phonetics Laboratory of the Department of Humanities.
Assessment Methods
Assessment will be conducted through a series of assignments distributed across the duration of the course. Students will also be required to participate in structured discussions and debates. Additionally, students will deliver presentations, which will serve as a component in demonstrating their ability to illustrate and discuss about the topics covered in the course.
Texts
• Daniel Jurafsky & James H. Martin (2023, online in open-access). Speech and Language Processing. (chapter 15 Chatbots and Dialogue Systems, chapter 16 Automatic Speech Recognition and Text-to-Speech). • Fabio Tamburini (2022). Neural Models for the Automatic Processing of Italian. Bologna: Pàtron Editore. (Part II The Automatic Analysis of Italian: chapter 8 Let the Speech Speak! Parte III Surfing on the next wave)
Contents
The techniques of speech recording, acoustic phonetics, speech processing, speech classification, ASR, and TTS will be explored in detail, both from a theorerical and from a practical perspective. The various types of filters, waveforms, spectrograms, automatic extraction of segmental and suprasegmental features, the Mel scale, Mel spectral coefficients, ASR systems based on Hidden Markov Model and those of the Deep Neural Networks type will be investigated. The laboratory activities will be dedicated to the use of the most modern techniques for elicitation, recording and analysis of speech (read and spontaneous). Various professional software will be used (e.g., Praat, MAUS, etc.) for forced alignment, segmentation and processing of audio material, feature extraction and acoustic analysis. Furthermore, students will learn to deveolp speech classification, ASR, and TTS systems in Python.
Course Language
English
More information
The course is offered by Collegio Ghislieri as part of the “Collegiale non residente” project and will be held from January 8 to 16, 2026. N.B. Attendance to all classes is mandatory.
Degrees
Degrees (2)
2 years
2 years
No Results Found
People
People
Teaching staff
No Results Found