Registration - Speech Synthesis Workshop 2027

First Call For Papers SSW 2027

The Speech Synthesis Workshop (SSW) is the main meeting place for research and innovation in speech synthesis, i.e. predicting speech signals from text input. Text-to-Speech (TTS) technology is a key component of numerous applications: speech-to-speech translation, digital assistants, conversational agents, social robots. While early research focused on basic intelligibility, contemporary systems now achieve remarkable naturalness. Current research frontiers include emotional expression, speaking style control, and efficient deployment for the world’s languages.

SSW welcomes contributions not only in the core TTS technology but also gathers researchers from contributing sciences: from phoneticians, phonologists, linguists, neuroscientists to experts of multimodal human-machine interaction.

Since 1990, SSWs are held every three years under the auspices of ISCA‘s special interest group SynSIG. In 2019, it was decided to have an SSW every two years, since the technology is advancing faster these days.

The 14th ISCA Speech Synthesis Workshop will be held in Maceió, Brazil, 6 (Mon)- 8 (Wed) Sept 2027. The workshop is a satellite event of the INTERSPEECH 2027 conference (São Paulo, Brazil).

Workshop topics

Papers in all areas of speech synthesis technology are encouraged to be submitted, including but not limited to:

Grapheme-to-phoneme conversion for synthesis
Text processing for speech synthesis (text normalization, syntactic and semantic analysis)
Segmental-level and/or concatenative synthesis
Signal processing/statistical model for synthesis
Speech synthesis paradigms and methods; articulatory synthesis, parametric synthesis etc.
Prosody modeling and generation
Expression, emotion and personality generation
Voice conversion and modification, morphing
Concept-to-speech conversion speech synthesis in dialog systems
Avatars and talking faces
Cross-lingual and multilingual aspects for synthesis
Applications of synthesis technologies to communication disorders
TTS for embedded devices and computational issues
Tools and data for speech synthesis
Quality assessment/evaluation metrics in synthesis
Singing synthesis
Synthesis of non-human vocalisations
End-to-end text-to-speech synthesis
Direct speech waveform modelling and generation
Speech synthesis using non-ideal data (‘found’, user-contributed, etc.)
Natural language generation for speech synthesis

Special Theme: Speaking in sub-tongues: embracing culture, heritage and diversity in Brazilian accents

Since the theme of this edition is Speaking in sub-tongues: embracing culture, heritage and diversity in Brazilian accents, the following key topics are also proposed:

Preserving Regional Accent Markers: How to generate natural speech that retains the phonetic and phonological identity of a specific regional accent, including front end techniques such as accent aware G2P conversion that reflects how words are articulated locally.
Accent Intensity Control: Modeling approaches that allow adjusting the weight of an accent, from subtle to heavy, whether at the phrase level, per phoneme, or as a continuous control parameter.
Prosody Control for Dialectal Variation: Dialects, going beyond segmental features to capture the musicality of regional speech.
Low Resource Accent Modeling: Transfer learning, fine tuning, and few shot strategies to create convincing regional accents when only minutes or a handful of samples of labeled speech are available.
Evaluating Accented Speech: Naturalness, Identity, and Trade offs: Analyses of the interplay between accent fidelity, overall naturalness, and speaker identity preservation. What is gained or lost when we prioritize one dimension over another.

Important dates

April 01, 2027	Paper Submission System Opens
April 20, 2027	Final paper submission
June 21, 2027	Registration opens
June 21, 2027	Notification of acceptance
July 04, 2027	Camera-ready
Aug 29 - Sep 2, 2027	INTERSPEECH 2027, São Paulo, Brazil
September 6-8, 2027	SSW14, Maceió, Alagoas, Brazil
September 9, 2027	Blizzard Challenge 2027, Maceió, Alagoas, Brazil

Submission Instructions

The workshop program will consist of a single track with three invited talks, oral and poster presentations.

The deadline for submissions is Sunday, April 20, 2027, 23:59, Anywhere on Earth. Make sure that your paper complies with all instructions and specifications in the paper kit and with the ISCA Code of Ethics for Authors.

Submission Guidelines

1. Language

English

2. Paper Length

Regular Papers: Up to 4 pages for the main content, plus up to 2 additional pages reserved exclusively for references, acknowledgments, and disclosure of any generative AI tool use. Max 4+2 Pages
Long Papers: Up to 8 pages for the main content, plus up to 2 additional pages reserved exclusively for references, acknowledgments, and disclosure of any generative AI tool use. Max 8+2 Pages

3. Abstract

A maximum of 1,000 characters, including spaces.

4. Double-Blind Review

Interspeech 2027 uses a double-blind review process. Authors must prepare an anonymized version for review and a camera-ready version for final submission. Please follow all instructions in the Paper Kit, which includes complying with the submission policy.

5. File Type

Manuscripts must be submitted in PDF format only.

6. Layout and Templates

Authors must use the Interspeech 2027 Paper Kit, available on the conference website and through the Overleaf Template Gallery. Do not reuse files from previous Interspeech templates, formatting and policy updates have been made for 2027.

Note: For MS Word, both “review” and “camera ready” templates are available in the Paper Kit. For LaTeX, there are different document classes for “review” and “camera ready” documents. Please select the appropriate one.

Submission Platform

CMT Submission Portal

Papers should be submitted via the CMT platform:

This section is currently under development. Detailed information will be published here as soon as updates become available.

Double-blind reviews (authors and reviewers are anonymous) will be performed. Please ensure that the author names and affiliations are not visible in the submitted document and the submission does not contain any information (references, repository addresses, etc.) that could be used to identify the authors.

Please refrain from publishing the submissions on online platforms, such as pre-print repositories, during the review process.

Full papers will receive a DOI and be published in the ISCA online archive.

Note: For complete information about SSW 2027, please visit the website https://nilc-nlp.github.io/ssw-2027/.

Speech Synthesis Workshop 2027