37.3265, Confs: Workshop at PAPE 2027: Creativity & Acoustic Variability in Human and Generative Speech (Portugal)

[The LINGUIST List]

LINGUIST List: Vol-37-3265. Mon Oct 12 2026. ISSN: 1069 - 4875.

Subject: 37.3265, Confs: Workshop at PAPE 2027: Creativity & Acoustic Variability in Human and Generative Speech (Portugal)

Moderator: Steven Moran (linguist at linguistlist.org)
Managing Editor: Valeriia Vyshnevetska
Team: Helen Aristar-Dry, Daniel Swanson
Jobs: jobs at linguistlist.org | Conferences: callconf at linguistlist.org | Pubs: pubs at linguistlist.org

Homepage: http://linguistlist.org

Editor for this issue: Valeriia Vyshnevetska <valeriia at linguistlist.org>

================================================================


Date: 09-Oct-2026
From: Farhat Jabeen [farhat.jabeen at uni-bielefeld.de]
Subject: Workshop at PAPE 2027: Creativity & Acoustic Variability in Human and Generative Speech


Workshop at PAPE 2027: Creativity & Acoustic Variability in Human and
Generative Speech
Short Title: CAViS
Theme: Phonetics and Phonology

Date: 17-Jun-2027 - 17-Jun-2027
Location: Lisbon, Portugal
Meeting URL:
https://labfon.letras.ulisboa.pt/PAPE2027/workshop-cavis.html

Linguistic Field(s): Phonetics; Phonology

Submission Deadline: 13-Nov-2026

Summary Description:
This session aims to explore the potential of generative speech models
(GSMs) as an alternative data source for phonetic research,
particularly in prosodic analysis. Current GSMs are claimed to
successfully generate output at par with human speech (Chen et al.,
2024). Despite growing interest in GSMs, their use for prosodic
research remains underexplored and has yielded mixed findings. For
example, Minixhofer et al. (2025) showed that GSMs can generate speech
comparable to human production. In contrast, Hu et al. (2024) and
Vigneaux and Cole (2026) reported intonational variability in
GSM-generated speech that is unattested in human data. However, these
studies assessed different aspects: Minixhofer et al. focused on
output quality, whereas Hu et al. and Vigneaux and Cole evaluated
specific linguistic phenomena. These findings highlight the need for a
systematic investigation of GSMs’ ability to model speech.
Recent non-deterministic models are particularly promising because
they can capture variability rampant in natural speech. Moreover,
Jabeen and Lai (2026) demonstrated that GSMs differ in their ability
to model intonation depending on the phenomena under investigation
(e.g., accentuation versus deaccentuation). While GSMs may therefore
represent a valuable resource for phonetic research, it remains
unclear how they can be most effectively leveraged.
As most of the recent work in this field is based on (American)
English (Kuang et al., 2026; Salem and Levitan, 2026; Sheppard, 2026),
we particularly encourage submissions on under-resourced languages.
This workshop will address the following questions:
- How can Generative Speech Models (GSMs) be used most effectively for
phonetic research, including the study of segmental processes,
intonation, voice quality, timing, and rhythm etc.?
- Which linguistic phenomena lend themselves most readily to speech
generation by GSMs?
- How should tasks and models be selected based on available
functionalities and levels of control?
- To what extent are GSMs a feasible resource for low-resource
languages?
- What strategies can enable us to move beyond generic output patterns
when evaluating the originality and communicative success of generated
speech?
Submission Information:
We invite abstracts related to the above-mentioned areas of interest.
Single page (A4) abstracts should be written in English and submitted
as a pdf file here. Examples, figures, tables, and references may be
added on the second page. All the abstracts will be peer-reviewed.



------------------------------------------------------------------------------

********************** LINGUIST List Support ***********************
Please consider donating to the Linguist List, a U.S. 501(c)(3) not for profit organization:

https://www.paypal.com/donate/?hosted_button_id=87C2AXTVC4PP8

LINGUIST List is supported by the following publishers:

Australian Linguistics Society https://als.asn.au/Home

Bloomsbury Publishing http://www.bloomsbury.com/uk/

Cambridge University Press http://www.cambridge.org/linguistics

Cascadilla Press http://www.cascadilla.com/

De Gruyter Brill https://www.degruyterbrill.com/?changeLang=en

Edinburgh University Press http://www.edinburghuniversitypress.com

European Language Resources Association (ELRA) http://www.elra.info

John Benjamins http://www.benjamins.com/

Language Science Press http://langsci-press.org

MDPI Languages https://www.mdpi.com/journal/languages

MIT Press http://mitpress.mit.edu/

Multilingual Matters http://www.multilingual-matters.com/

Narr Francke Attempto Verlag GmbH + Co. KG http://www.narr.de/

Netherlands Graduate School of Linguistics / Landelijke (LOT) http://www.lotpublications.nl/

Peter Lang AG http://www.peterlang.com

SIL International Publications http://www.sil.org/resources/publications

University of Toronto Press http://www.utpjournals.com/


----------------------------------------------------------
LINGUIST List: Vol-37-3265
----------------------------------------------------------