The 3rd Workshop on Automatic Simultaneous Translation

Call for Papers for the 3rd Workshop on Automatic Simultaneous Translation

We are happy to announce the call for papers for the 3rd workshop on Automatic Simultaneous Translation co-located with NAACL 2022.

More details on the workshop, including topics of interest, important deadlines, instructions for authors and shared tasks are below:

Simultaneous translation, which performs translation concurrently with the source speech, is widely useful in many scenarios such as international conferences, negotiations, press releases, legal proceedings, and medicine. It combines the AI technologies of machine translation (MT), automatic speech recognition (ASR), and text-to-speech synthesis (TTS) and is rapidly becoming a cutting-edge research field. As an emerging and interdisciplinary field, simultaneous translation faces many great challenges.

To promote the development in this field, we held the first and the second workshop on simultaneous translation at ACL 2020 and NAACL 2021 respectively. We attracted 94 registered participants in 2020 and 96 registered participants in 2021.The 3rd workshop will bring together researchers and practitioners in machine translation, speech processing, and human interpretation, to discuss recent advances and open challenges of simultaneous translation.

Possible topics include but are not limited to:

• Simultaneous translation paradigms: traditional pipeline (ASR-MT-TTS) or end-to-end (speech-speech);
• Translation models: robust, high quality, and low latency;
• Data resources: large and high-quality corpora for training simultaneous translation systems.
• Evaluation methods: metrics to evaluate the translation quality and time latency;
• Computer Aided Interpretation (CAI): improve the efficiency and quality of human interpreters.

The workshop will also explore challenges for developing practical simultaneous translation systems and will provide opportunities for participants to show their products.

We introduce three submission tracks in this workshop:

• Research track for research papers;
• Highlights track for research papers accepted elsewhere (non-archival submission);
• Shared task track for systems description papers in the shared task.

Important Dates

Shared Task

• Registration and Release of data: March 7th, 2022
• End of registration: April 30th, 2022
• System Submission: May 1st, 2022
• System Description Due: May 14th, 2022
• Notification of Acceptance: June 1st, 2022
• Camera-ready Papers Due: June 16th, 2022

Papers

• Submission Deadline: May 1st, 2022 (all the three tracks: research, non-archival highlights and shared task)
• Notification of Acceptance: June 1st, 2022
• Camera-ready Papers Due: June 16th, 2022
• Workshop: TBD, 2022

Instruction for authors

Our AutoSimTrans Workshop follows the submission policy and format policy of NAACL 2022 Submission Guidelines that are posted under PAPER SUBMISSION INFORMATION on ACL Rolling Review page with respect to author guidelines, templates, double blind review, ethics and referencing prior work.

There are two formats: long papers (8 pages content plus unlimited references) and short papers (4 pages contect plus unlimited references). Authors can also upload a separate appendix of unlimited length, which is optional for the reviewers. The final versions of both long and short papers will be given one extra page of content to address reviewers comments.

Non-Archival Submission

We allow dual submission of work to be presented in the Highlights track. Please submit to the relevant track. These papers will not be included in the official proceedings.

System Description Paper Submission

For those who participate in the shared task, you are encouraged (though not required) to submit a system description paper of 4 to 8 pages of content plus unlimited references to AutoSimTrans to describe your system.

Please submit via OpenReview.net.

Shared Task Submission

We are organizing a simultaneous translation shared task in two directions with different input modes: (a) Chinese-to-English with streaming transcription as input, (b) Chinese-to-English with speech audio as input, and (c) English-to-Spanish with streaming transcription as input. There are three competitions in total. We will provide open datasets and an evaluation environment. Participants are encouraged (though not required) to submit their system description papers and present them at the Workshop.

For more details, please refer to the Shared Task information at https://autosimtrans.github.io/shared.