Albayzin Evaluation
IberSpeech-RTVE 2018 Challenge

Evaluation Data

Following the success of previous “ALBAYZIN” technology evaluations since 2006, this year ALBAYZIN evaluations shall focus around multimedia analysis of TV broadcast content. Under the framework of a newly created Cátedra RTVE de la Universidad de Zaragoza, we launch the IberSPEECH-RTVE 2018 Challenge. The Corporación Radiotelevión Española (RTVE) will provide participants with an annotated TV broadcast database and the necessary tools for the evaluations. The evaluation is supported by the Spanish Thematic Network on Speech Technology (RTTH) and Cátedra RTVE de la Universidad de Zaragoza and is organized by ViVoLab Universidad de Zaragoza . The evaluation will be conducted as part of the Iberspeech 2018 conference to be held in Barcelona from 21 to 23 November 2018.

RTVE2018 Database

RTVE2018 database is a collection of whole TV shows drawn from diverse genres and broadcast by the public Spanish National Television from 2015 to 2018. There are a total of 569 hours and 22 minutes of audio. About 460 hours are provided with the subtitles and about 109 hours have been human-revised transcribed. The database has been divided in 4 partitions: a train one, two development partitions dev1, dev2 and a test partition. Additionally, the database includes a set of text files extracted from all the subtitles broadcast by the RTVE 24H Channel during 2017.

  • More information can be found in the RTVE2018 database description report

RTVE2018 Database License

The data is available to the IberSPEECH-RTVE 2018 Challenge participants only and subject to the terms of a licence agreement with the RTVE. To download the RTVE data, you will need to sign this data license and return it to the IberSpeech-RTVE team.

  • Download license agreement, please contact organizers.
  • Please, return a scanned copy of the license with your signature to lleida@unizar.es , with the subject: RTVE2018 database license

Important information about license extension:

At the end of the challenge, the licensee shall send a written document to RTVE guaranteeing that all the material has been destroyed, unless he or she has applied for a license extension. RTVE will authorize the use of the contents released for the call IBERSPEECH-RTVE Challenge 2018, for its use in research works, to all those participants who request it. The authorization will be valid for three years from the date of the public communication of the results of the Challenge 2018. After this period, if necessary, an extension may be requested for the same use. Applications must be sent by mail to
RTVE
Dir. FONDO DOCUMENTAL RTVE
Avda. Radiotelevisión, 4
28223 Pozuelo de Alarcón
Madrid
España
The application must indicate user, purpose of research and period of use.

RTVE2018 Database Download

Once we have processed data license, you will receive an email with the instructions to download the database. If you haven't received, please contact us.