How well do these systems capture prosodic variation across languages and language varieties?
Dates: Wednesday, 7 and Thursday, 8 August
Scheduled during the poster sessions, posters 64-68.
Prosody Visualisation Challenge 2 (PVC2) is the second event in the series following the success of the first challenge held at the Speech Prosody conference in Poznan in 2018. The main aims of the challenge are to bring together researchers working on various aspects of prosody and to show and share multiple approaches to prosody visualisation.
PVC2 aligns with the themes for ICPhS 2019 “Endangered Languages, and Major Language Varieties” and considers systems for the representation of prosody that best capture diversity in prosodic phenomena across well-described language varieties (i.e. English) and typologically distinct, less documented, and/or endangered languages.
The following systems have been selected for the challenge. There will be a poster display of these entries in the poster sessions on Wednesday and Thursday at the Congress. Sample sound files will also be available to listen to:
Paper 1 (poster number 66)
Using jaw movement patterns to visualize prosody
Donna Erickson1, Ting Huang1, Caroline Menezes2 and Shigeto Kawahara3
1Haskins Laboratories, 2 University of Toledo, Ohio, 3 Keio University, Japan
Paper 2 (poster number 65)
Visualising macro-rhythm as an indicator of prosodic typology
Catalina Torres and Kathleen Jepson
University of Melbourne, Australia
Paper 3 (poster number 68)
Can you draw me a question?
Aviad Albert, Francesco Cangemi and Martine Grice
IƒL-Phonetik, University of Cologne, Germany
Submit your vote!
All delegates at ICPhS 2019 are invited to view the poster display for these systems, and then submit their vote for the best system online using the link below. The winning system will be announced at the closing ceremony of ICPhS:
‘Prosody’ is often a mystery to those outside it, and it is not always easy to describe prosody – the phenomena variously called intonation, timing, voice quality, stress, prominence, juncture, vocal effort, and so on – to non-prosodists and to other researchers. We can play or perform examples. But there are good reasons, both practical and scientific, to describe prosody more abstractly, in ways that allow a visual summary of patterns.
To date, a number of diverse approaches and systems to capture prosody are available. However, these are often hard for outsiders to interpret without any prior knowledge, and at times, even for those who specialise in the study of prosody. Also, direct comparisons showing similarities and differences, advances and limitations between the systems are not always available. In addition, many existing systems focus on phenomena related to intonation (f0 contour and f0 range, prominences, prosodic structure, etc.) and may not capture both timing characteristics and intonational properties of a language in a single system, or have been developed through extensive work on European languages (such as English), with a limited application to less documented languages.
PVC2 invited all to bring their systems of prosody visualisation for a comparative display. This helps everyone to know what systems are available and how they compare when looking at prosodic variation across languages and dialects.
The challenge is being held in the format of a poster session as part of the ICPhS 2019 program. The winner of the challenge will be chosen by votes cast by the session attendees.
Challenge guidelines (submissions now closed)
- To take part in the challenge, please submit an abstract (not more that 2 pages) and a title by email (email@example.com). Deadline for submission is July 15, 2019.
- Each accepted entry will be presented via an illustrative poster, as well as a set of analysed examples in a digital form, available on a laptop at the time of poster session.
- Poster size: A0 (841mm wide x 1189mm high or 33.1in by 46.8in), portrait orientation.
- There is no associated requirement for a paper in the proceedings, nor is there a novelty requirement for the presentation system. (In contrast: existing descriptions of the system used – in a journal article, conference proceedings, technical report, or website – will increase the likelihood of a Challenge submission being accepted!)
- At least one author of the submission must be listed as an author on another paper and be registered for the Congress.
- Entrants may base their analysis on any languages/language varieties they are working on. The challenge is to select a good set of samples in which your visualisation system best captures the contrasts/similarities across languages and language varieties, and to explain in your poster how these can be seen. There is thus no common audio requirement – use your own audio!
- Please submit your sample recordings with your abstract. Please also include below your abstract a statement to certify that you have appropriate ethics permission from the speakers in the recordings to use them for this purpose. Note the recordings will be available for people attending the Congress to listen to during the poster session.
- There is no “gold standard” and no scale for scoring the systems. All entries will be on display during the two poster sessions. Congress attendees will be provided with a response form on which they can rate and comment on the entries.
- The winner of the challenge will be announced at the closing ceremony.