Advanced Lecture, 2 SWS, 3 LP/ECTS, LSF #134460
MSc LST/LCT; MSc Inf; MSc CuKT
Thu 10.15-11.45 AM
Building E1.3, Room 0.03.1 (Hörsaal III)
Exam inspection is possible on Thu Oct 27, 1-2 PM, C7.2/1.01 (exam office, Cristina Deeg), but only after registration by email.
Formally speaking, none, but some background in speech science, signal processing, or computational linguistics will be advantageous.
Written exam at the end of the summer semester. Registration via LSF-POS or, if this is not possible, by email to the lecturer.
Speech synthesis is an essential component of any system relying on intuitive human-machine communication. Speech synthesis systems are also used in phonetic research to gain further insight into speech production and acoustic properties of speech. This advanced course offers an introduction to text-to-speech (TTS) synthesis systems and strategies. Various approaches to speech synthesis are presented, including formant synthesis, concatenative synthesis, and state-of-the-art unit selection and statistical (HMM and DNN based) synthesis, as well as recent end-to-end approaches. Linguistic text analysis and natural language processing modules typically found in TTS systems are covered as well.
Contact:
| Date | Topic | Slides | Background |
| 21.04. | Introduction: history, methods | pdf / pptx (incl. audio) | Taylor 2009, ch. 1 and 2 |
| 28.04. | TTS system components | Taylor 2009, ch. 7.1-7.2 | |
| 05.05. | Formant synthesis | pdf / pptx (incl. audio) | Taylor 2009, ch. 13 |
| 12.05. | Diphone synthesis | pdf / pptx (incl. audio) | Taylor 2009, ch. 14 |
| 19.05. | Unit Selection synthesis | pdf / pptx (incl. audio) | Taylor 2009, ch. 16 |
| 26.05. | (no class) | ||
| 02.06. | Duration modeling | Taylor 2009, ch. 9 | |
| 09.06. | Intonation modeling | Taylor 2009, ch. 9 | |
| 16.06. | (no class) | ||
| 23.06. | Linguistic text analysis I | Taylor 2009, ch. 7.4, 8.3-8.5 Möbius 2001, ch. 3-6 |
|
| 30.06. | Linguistic text analysis II | (see 23.06.) | |
| 07.07. | Parametrical and NN-based synthesis Taught by S. Le Maguer |
pdf /
zip (incl. audio) (start with index.html) |
Taylor 2009, Ch. 15 Dale: Voice synthesis business 2022 |
| 14.07. | TTS system evaluation Taught by S. Le Maguer |
Taylor 2009, ch. 17.2 Wester 2015: read at least first three sections |
|
| 21.07. 10:15-11:45 |
Written exam E1.3/0.03.1 (Hörsaal III) |
Registration for exam by July 14, 2022 via LSF/HISPOS or else by email to Prof. Möbius |
|
| 19.10. 14:15-15:45 |
Written exam (resit) B3.1/0.14 (Hörsaal I) |
Registration for exam by Oct 12, 2022 via LSF/HISPOS or else by email to Prof. Möbius |
BibTex entries of all references (books, papers, URL)