| Contents | Teachers | Prerequisites | Enrollment | Literature | Requirements | Schedule | Teams, Participants, Topics, and Student Presentation Schedule |

The seminar is devoted to changing current topics in audio signal processing. In fall 2007, the title of the seminar will be signal processing for sound synthesis. The seminar will focus on digital signal processing methods that can be used for imitating the sounds produced by musical instruments and others sound sources. The language of the seminar is English.

Teachers

• The seminar is coordinated by professor Vesa Välimäki and M.Sc. Jyri Pakarinen.
• Contact: S-89.3580@acoustics.hut.fi

It is required to know basics of acoustics and digital signal processing, such as the courses S-89.2300 Sound and Voice Technology (that replaces S-89.300) and T-61.3010 Digital Signal Processing and Filtering, or their equivalents.

Please register yourself for this course using WebTOPI.

Parts of textbooks and papers from journals and conferences.

• DAFx Web Page (first select the year of the conference; papers accessible through the proceedings-link)
• AES On-line Electronic Library (direct link that works from the campus). Papers can also be obtained from the Acoustics Laboratory Library.
• Google Scholar : great tool for searching academic publications (shows patents, too).
• Google Patents : great tool for searching US patents.
• Signal Processing for Sound Synthesis, IEEE Signal Processing Magazine, special section 24(2), March 2007.

You can choose either "S-89.3580 Audio Signal Processing Seminar" (3 cr) or "S-89.4820 Postgraduate Course in Audio Signal Processing" (8 cr). There are differences in requirements and the amount of work between the two courses.

• Requirements for S-89.3580 Audio Signal Processing Seminar (3 cr)
• Requirements for S-89.4820 Postgraduate Course in Audio Signal Processing (8 cr)

Participation in the first lecture and all the tutorial lectures is part of the course requirements. The seminar will be held at the TKK Electrical and Communications Engineering Building, Room E111 on Wednesdays 9.15-12.00. Note: the tutorial lecture titles below are tentative, i.e. they might be subject to change.

• 12.09.2007, 09.15-12.00, Jyri Pakarinen, Introduction to the seminar topic. Dr. Balazs Bank (visiting postdoc researcher at TKK from Budapest Univ. of Technology and Economics, Hungary), Introduction to physical modeling of musical instruments
• 19.9.2007, Prof. Vesa Välimäki, Virtual analog synthesis
• 26.9.2007, Dr. Cumhur Erkut, Block-based physical modeling for digital sound synthesis
• 3.10.2007, Jyri Pakarinen, New topics in guitar modeling, part 1: Tube amplifier modeling, part 2: Virtual Slide Guitar.
• 10.10.2007 Jyri Pakarinen and Vesa Välimäki, Guidelines for presenting and writing scientific text. David Yeh, (CCRMA, Stanford University, USA), Physical and behavioral circuit modeling of the SP-12 sampler
• 17.10.2007 No tutorial lecture. The course personnel will be at E111 9.15-10 if you wish to discuss your seminar topic, etc.
• 24.10.2007 First student presentation: Synthesis control and parameter estimation
• 31.10.2007 No lecture, exam period.
• 7.11.2007, 9:15-12:00 (E111) and 13:00-16:00 (Piano-bar), Prof. Davide Rocchesso (IUAV University of Venice, Department of Art and Industrial Design, Italy), Sonic interaction design and sound synthesis.
• 14.11.2007, Dr. Mikael Laurson (Sibelius Academy) Score-based control of sound synthesis: the guqin as a case study.
• 21.11.2007 Second student presentations: Singing-voice synthesis and transformations.
• 28.11.2007 Third student presentations: Filters and modulation-based sound synthesis
• 5.12.2007 Fourth student presentations: Wave-based synthesis
• 13.12.2007, Prof. Augusto Sarti (Politecnico di Milano, Italy), "Real time path tracing for applications of interactive acoustic rendering"

The students will discuss the seminar topics and plan the contents and coverage of their papers and presentations in small teams. However, within the team, each student will write his/her seminar paper and presentation slides individually. For each team, a team leader will be chosen (the bolded name on the table below). The team leader's task is to compile the team member's presentation slides into a single slideshow. Also, the team leader will email the draft versions of the papers and slides of his/her team to S-89.3580@acoustics.hut.fi before the deadlines below. See the requirements section for more detailed information.

Team Name Topic v1.0 due v2.0 due v3.0 due Presentation Advisor 1. Synthesis control and parameter estimation Mika Ristimäki Gestural control of sound synthesis 3.10. 10.10. 17.10 24.10. Jyri Sharman Jagadeesan Synthesis parameter extraction I Hannes Gamper Synthesis parameter extraction II 2. Singing-voice synthesis and transformations Jussi Pekonen Singing-voice synthesis using source filter methods 31.10. 7.11. 14.11. 21.11. Vesa Jukka Kautto Singing-voice synthesis using spectral modeling Marko Hiipakka Singing-voice transformations 3. Filters and modulation-based sound synthesis Heidi-Maria Lehtonen Instrument body modeling 7.11. 14.11. 21.11. 28.11 Vesa Jari Kleimola Dispersion modulation Azadeh Haghparast Fractional delay filters in sound synthesis Jaakko Kestilä Modulation-based synthesis methods 4. Wave-based synthesis Antti Jylhä Wave digital filters 14.11. 21.11. 28.11 5.12. Jyri Kimmo Kokki Digital waveguide mesh Hugo Monreal Wave-based modeling of the violin Kari Kivisalo Modeling of tension modulation in strings


Here are briefly the tasks required for the deadlines above. For a more thorough description, see the requirements section.
• V1.0: At least 3 weeks before your presentation, send a plan for your paper (preliminary table of contents and list of references) to your seminar advisor (Vesa or Jyri). If you need assistance at this step, talk to your advisor well before the deadline at the end of the lectures.
• V2.0: Then, prepare the second draft of the text (1/2 ready) and send it to your advisor, at least 2 weeks before your scheduled talk.
• V3.0: One week before the talk at the latest, you must send the full version of the paper, along with a draft of your presentation slides to Vesa or Jyri. The team leaders will forward the paper's to all team member's and you will start reading and commenting them.
• Two days before the presentation: Send the ready slides to Vesa for final checking and give your feedback concerning the other team member's papers.
• Before the presentation, the final paper must be copied to everybody.