UnivIS
Informationssystem der Friedrich-Alexander-Universität Erlangen-Nürnberg © Config eG 

Seminar Digital Pathology using Speech Language Processing (Sem.DPSLP) (Sem.DPSLP)2.5 ECTS
(englische Bezeichnung: Seminar Digital Pathology using Speech Language Processing (Sem.DPSLP))

Modulverantwortliche/r: Kubilay Can Demir, Tobias Weise
Lehrende: Seung Hee Yang


Startsemester: WS 2021/2022Dauer: 1 SemesterTurnus: halbjährlich (WS+SS)
Präsenzzeit: 30 Std.Eigenstudium: 45 Std.Sprache: Englisch

Lehrveranstaltungen:

  • The task of automatic speech recognition translates spoken language into text. It is an interdisciplinary subfield of computer science and computational linguistics that develops methodologies and technologies. The participants will have lectures in speech recognition, speech production, analysis, filter bank, cepstrum, feature extraction for ASR. linguistic categories for speech recognition, and neural networks for acoustic modelling and language modelling. The lectures will be accompanied by student presentations on the recent research papers published in INTERSPEECH and ICASSP.

    • Interpretation and Analysis of Neural and Muscle Signals
      (Seminar, 2 SWS, Alessandro Del Vecchio et al., Zeit n.V., Zoom-Meeting)
    • Deep Learning Exercises
      (Übung, 2 SWS, Zijin Yang et al.)

Inhalt:

The task of automatic speech recognition translates spoken language into text. It is an interdisciplinary subfield of computer science and computational linguistics that develops methodologies and technologies. The participants will have lectures in speech recognition, speech production, analysis, filter bank, cepstrum, feature extraction for ASR. linguistic categories for speech recognition, and neural networks for acoustic modelling and language modelling. The lectures will be accompanied by student presentations on the recent research papers published in INTERSPEECH and ICASSP.

Lernziele und Kompetenzen:


Verstehen
The learners will gain understanding of automatic speech recognition.
Anwenden
The learners will be able to build their own speech recogniser.
Analysieren
The learners can break down the problem of automatic speech recognition in terms of algorithms and linguistics.

Literatur:

[1] Jurafsky, Dan. Speech & language processing. Pearson Education India, 2000. [2[ Yu, Dong, and Li Deng. Automatic Speech Recognition. Springer london limited, 2016. [3] Jurafsky, Dan, and Christopher Manning. "Natural language processing." Instructor 212.998 (2012): 3482. [4] Haykin, Simon. Neural networks and learning machines, 3/E. Pearson Education India, 2010.


Weitere Informationen:

Schlüsselwörter: Automatic Speech Recognition, spoken language processing

Studien-/Prüfungsleistungen:

Seminar Digital Pathology using Speech Language Processing (Sem.DPSLP) (Prüfungsnummer: 882806)

(englischer Titel: Seminar Digital Pathology using Speech Language Processing (Sem.DPSLP))

mündliche Prüfung, Dauer (in Minuten): 30, benotet
Prüfungssprache: Englisch

Erstablegung: WS 2021/2022, 1. Wdh.: SS 2022
1. Prüfer: Seung Hee Yang,2. Prüfer: Seung Hee Yang

UnivIS ist ein Produkt der Config eG, Buckenhof