DT2112 Speech Technology 7.5 credits

The course Speech Technology deals with research and development in speech technology and also povides basic insights in speech, language and hearing. The course contains segments concerning:

• Linguistic theory and phonetics
• Speech production by humans (speech physiology and acoustics) and computers (text-to-speech synthesis)
• Speech perception by humans (hearing and psycholinguistics) and computers (speech recognition)
• Multimodal dialogue systems for human-computer interaction with speech and vision
• Practical speech technology studies and experiments
• Methods for evaluation of and experimentation in speech technology
• Data collection for speech technology research and development

After passing the course, the student should be able to

• describe speech from an acoustic, phonetic, and linguistic perspective
• explain how computers recognize speech with statistical methods, and evaluate the recognition results
• describe and judge different methods used to produce speech with computers
• analyze speech-driven dialogue systems with respect to application, components, functionality and user aspects
• describe how evaluation of speech technology systems work and describe the special requirements that are posed by this type of system
• design and describe data collections for speech technology research and development
• give an account of available state-of-the art speech technology and exemplify the current speech research on e.g., mobile systems and IT applications
• apply the theoretical knowledge in small-scale speech technology projects in teams with participants from different backgrounds.

This course brings together students with varying levels of prior knowledge relevant to speech technology, as defined by the specific entry requirements set out below.

• Knowledge and skills in programming, 6 credits, may, for example, have been acquired through a completed course in programming for technologists such as DD1337/DD1310-DD1319/DD1321/DD1331/DD1333/DD100N/ID1018/ID1022 (KTH) or programming for language technology/language informatics such as LIS020 (Stockholm University)/5LN429 (Uppsala University).
  
  
• Knowledge in any of the following areas that are relevant to the speech technology taught in this course:
  – algorithms and data structures, 6 credits, equivalent to completed course DD1338/DD1320-DD1327/DD2325/ID1020/ID1021 (KTH)
  – robotics, 7.5 credits, equivalent to completed course DD2410 (KTH)
  – human-computer interaction or interaction design, 7.5 credits, equivalent to completed course DH1622/DM2601/DH2628/CH2010 (KTH)
  – language technology/language informatics, 22.5 credits, equivalent to completed course or active participation (registration) in at least three of the following courses:
  LIS030 Analysis and Processing of Speech, 7.5 credits, Stockholm University
  LIS010 The Structure of Language, 7.5 credits, Stockholm University
  LIS040 Statistics and Machine Learning, 7.5 credits, Stockholm University
  5LN445 Mathematics for Language Technologists, 7.5 credits, Uppsala University
  5LN710 Natural Language Processing, 7.5 credits, Uppsala University
  5LN715 Advanced Programming for Language Technologists, 7.5 credits, Uppsala University
  
  
• Skills in English equivalent to English B/6, according to the Swedish upper secondary school system.

Proficiency in mathematics and signal analysis corresponding to the first 2.5 years of Degree Programme in Media Technology.

• TENA - Home exam, 3.5 credits, grading scale: A, B, C, D, E, FX, F
• PR01 - Project and report, 3.5 credits, grading scale: A, B, C, D, E, FX, F
• LABB - Research laboratory assignment, 0.0 credits, grading scale: A, B, C, D, E, FX, F
• LABA - Laboratory assignments, 0.5 credits, grading scale: P, F

Based on recommendation from KTH’s coordinator for disabilities, the examiner will decide how to adapt an examination for students with documented disability.

The examiner may apply another examination format when re-examining individual students.

If the course is discontinued, students may request to be examined during the following two academic years.

Only those who have passed the course component LABA and the project bid and project presentation in PRO1 may submit a project report in PRO1 and complete TENA and LABB.

The LABB exam is not mandatory but can give a higher grade on the course.

Active participation in scheduled activities.

Éva Székely

Joakim Gustafsson

• All members of a group are responsible for the group's work.
• In any assessment, every student shall honestly disclose any help received and sources used.
• In an oral assessment, every student shall be able to present and answer questions about the entire assignment and solution.

In this course, the EECS code of honor applies, see:
http://www.kth.se/en/eecs/utbildning/hederskodex