Granada, November 14-16, 2022

Albayzin 2022 Text and Speech Alignment Evaluation, task 2: Plenary session of the Basque Parliament

Over the last years, with the widespread adoption of data-intensive deep learning approaches to ASR, the semi-supervised collection of training data for ASR has gained renewed interest. The Internet is plenty of resources pairing speech and text. Sometimes the paired text is an accurate transcription of the spoken content, sometimes it is only a loose and/or partial transcription, or even a translation to some other language. Therefore, a text-to-speech alignment system able to discriminate accurately paired speech and text segments becomes a very valuable tool.
With that goal in mind, the proposed task will deal with a long audio file, including sections in two different languages (Spanish and Basque), corresponding to a plenary session of the Basque Parliament. The paired text has been extracted from session’s minutes and reflects only sections in Spanish. We aimed to focus on a single language and chose Spanish because most of the research groups aiming to participate in this evaluation would have ASR technology and resources for Spanish, but few would have them available for Basque.

The calendar for the Albayzín Text and Speech Synchronization evaluation is:

  • May 3rd, 2022: Registration opens
  • May 3rd, 2022: Release of the training and development data
  • September 4th, 2022: Registration deadline.
  • September 5th, 2022 Release of the evaluation data
  • October 16th, 2022: Submission deadline
  • October 24th, 2022: Results distribution to the participants
  • October 30th, 2022: Paper submission deadline
  • November 15th, 2022:Iberspeech 2022 Albayzin Evaluations special session in Granada
More details in the evaluation plan