VVZ API is not affiliated with ETH Zurich. Data might be outdated or incorrect. Please view the official ETHZ Vorlesungsverzeichnis for binding information.
Speech Processing II
Sprachverarbeitung II
Last Updated: 2026-02-05 15:19:51
Abstract
Interdisciplinary approaches to text-to-speech synthesis and speech recognition (continuation of course speech processing I).
Objective
In this course selected concepts and interdisciplinary approaches to text-to-speech synthesis and speech recognition are presented.
Content
Fundamentals of representation and application of linguistic knowledge: Introduction of the theory of formal languages, the Chomsky hierarchy, word analysis, finite state machines, parsing. Speech synthesis: Natural language analysis (for words and sentences), lexicon, grammar for natural language; generation of the abstract representation of pronunciation (phone sequence, accents, phrases). Additionally, the ETH text-to-speech system SVOX is discussed. Speech recognition: The statistical approach to speech recognition with hidden Markov models is detailed: Basic algorithms (forward, Viterbi and Baum-Welch algorithm), problems of implementation, HMM training, whole vs. subword modeling, isolated word recognition, continuous speech recognition, statistical and rule-based language models.
Resources
Lecture Notes
Yes (available at the assistence in ETZ D97.5)
General Information
- Language
- German
- Levels
- DS , MSC
- Frequency
- Yearly recurring
Examination
- Type
- session examination
- Mode
- oral 30 minutes
Course Components
| Type | Title | Time & Place | Hours |
|---|---|---|---|
| lecture with exercise | Sprachverarbeitung II |
|
4 h weekly |