Blizzard Challenge 2023 Workshop

Call for participation

The Blizzard Challenge 2023 Workshop is the culmination of the Blizzard Challenge 2023. Blizzard Challenge is an annual challenge to compare corpus-based speech synthesis on common databases and a large listening test. The aims of the workshop are to present the results from the listening tests and for participants in the Challenge to describe their systems.

Who can attend the workshop ?

The workshop is open to all and we encourage participation from anyone interested in speech synthesis. However, please follow the registration procedure below.

Who can submit a paper to the workshop ?

All participants in the Blizzard Challenge 2023 are required to submit a paper describing their entry. The paper submission instructions can be found at Blizzard Challenge 2023 Rules #PAPER.

Organizers of the Blizzard Challenge 2023

• Olivier Perrotin & Gérard Bailly (Université Grenoble Alpes)
• Simon King (University of Edinburgh)

Location and date

Venue: Same as SSW12 https://ssw2023.org/index.php/venue/

• Maison de la Création et de l’Innovation – MaCI
• 339 Av. Centrale, 38400 Saint-Martin-d’Hères
• Grenoble, France

Date: Tuesday 29th August 2023

Registration

• If you are registered for SSW12 then you may attend the Blizzard Challenge 2023 without further registration - simply bring your SSW12 badge with you.
• If you are attending the Blizzard Challenge 223 without registering for SSW12, please email blizzard-challenge-organisers@googlegroups.com to register.

Programme

• We will have Live Oral Presentation for all teams that physically attend the workshop. The duration of each presentation is 15 min (half presentation, half Q&A).
• We will watch Pre-recorded videos from all teams that cannot physically attend the workshop. The duration of each presentation is 6 min (3 min video + 3 min Q&A).

All times below are in Paris time (GMT+2).

• 09:00 - 10:15 | Welcome message and Summary of Blizzard Challenge 2023 (Presenter: Olivier Perrotin)

• 10:15 - 10:30 | Break

• 10:30 - 11:45 | System presentations: FastSpeech-based models (live - 5 * 15 min)
  • LIUM-TTS - Laboratoire d'Informatique Le Mans Université (LIUM)
  • GIPSA-lab - Univ. Grenoble Alpes, CNRS, Grenoble INP, France
  • IMS - University of Stuttgart, Institute for Natural Language Processing, Germany
  • MuLanTTS - Microsoft
  • Samsung TTS - Samsung Electronics HQ and Samsung Research China, Beijing

• 11:45 - 12:00 | System presentations : FastSpeech- and Tacotron-based models (remote - 2 * 6 min)
  • SCUT SCSE (remote) - South China University of Technology
  • FireRedTTS (remote) - Xiaohongshu Inc.

• 12:00 - 13:30 | Lunch

• 13:30 - 14:30 | System presentations: Tacotron-based models (live - 4 * 15 min)
  • AudioLabs - International Audio Laboratories Erlangen
  • TTS-Cube - Adobe Systems, SCC
  • La Forge - Ubisoft
  • DeepZen - DeepZen Ltd.

• 14:30 - 15:00 | Break

• 15:00 - 16:00 | System presentations: Stochastic models (live and remote)
  • Idiap - Idiap Research Institute, Martigny, Switzerland
  • BIGAI - Beijing Institute of General Artificial Intelligence
  • CASIA Speech (remote) - Institute of Automation, Chinese Academy of Sciences (no-show; no video provided)
  • Xiaomi-ASLP (remote) - Xiaomi AI Lab and Audio Speech and Language Processing Group (ASLP@NPU), Northwestern Polytechnical University (video provided but no-show for Q&A)
  • Fruit Shell (remote) - University of Chinese Academy of Sciences (no-show; no video provided)
  • 10AI (remote) - Beijing Yiling Intelligence Technology Co., Ltd.
  • IOA-ThinkIT (remote) - Institute of Acoustics of the Chinese Academy of Sciences

• 16:00 - 16:30 | Closing

Published proceedings

TBA