Blizzard Challenge 2015: Difference between revisions

From SynSIG
No edit summary
No edit summary
 
(28 intermediate revisions by 2 users not shown)
Line 1: Line 1:
DeitY: Gov. of India, Apple Inc.,  and Google Inc., have generously provided financial support to the Blizzard Challenge 2015.
[[Image:Deity_logo.png||frame| http://deity.gov.in/]]
[[Image:Deity_logo.png||frame| http://deity.gov.in/]]
[[image:Google_logo.JPG||frame|right| Google UK has generously provided financial support to the Blizzard Challenge 2015]]
== This Blizzard Challenge has finished ==
* Please do not try to register for this challenge - you are too late!
* The data are no longer available via this site
* However, related data may be available from https://www.iitm.ac.in/donlab/tts/index.php
* The remainder of this page is left as a record of the Challenge


== Read these first ==
== Read these first ==


* First, please read the [[media:Blizzard2014.pdf | Call for Participation in the Blizzard Challenge 2015]] in PDF format
* First, please read the [[media: Blizzard2015-CallforParticipation.pdf | Call for Participation in the Blizzard Challenge 2015]] in PDF format
* Then, read and agree to the [[Blizzard Challenge 2015 Rules]] before participating
* Then, read and agree to the [[Blizzard Challenge 2015 Rules]] before participating
* You should only register for the challenge if you actually plan to submit an entry to the challenge
* You should only register for the challenge if you actually plan to submit an entry to the challenge
Line 12: Line 23:


2. Complete the license for Indian language data, which can be accessed via http://wissap.iiit.ac.in/blizzard2015/
2. Complete the license for Indian language data, which can be accessed via http://wissap.iiit.ac.in/blizzard2015/
3. Optionally, complete the license for the English pilot data, which can be accessed via http://www.cstr.ed.ac.uk/projects/blizzard/2015/usborne_blizzard2015


== Data download ==
== Data download ==


* Indian languages: About 4 hours of speech data in each of three Indian languages (Hindi, Tamil and Telugu), and about 2 hours of speech data in each of other three Indian languages (Marathi, Bengali and Malayalam) recorded by native professional speakers in high quality studio environments. Text is provided in UTF-8 format. No other information, such as segment labels, is provided.
=== Indian Languages ===
 
* About 4 hours of speech data is available in each of three Indian languages (Hindi, Tamil and Telugu), and about 2 hours of speech data in each of other three Indian languages (Marathi, Bengali and Malayalam), all recorded by native professional speakers in high quality studio environments. Text is provided in UTF-8 format. No other information, such as segment labels, is provided.


* These speech databases are provided by the group of institutions: IIT-Madras, IIIT-Hyderabad, SSNCE, CDAC Trivandrum, CDAC Mumbai, CDAC Kolkata.
* These speech databases are provided by the group of institutions: IIT-Madras, IIIT-Hyderabad, SSNCE, CDAC Trivandrum, CDAC Mumbai, CDAC Kolkata.


* You will be sent a username and password to download the data after accepting the license. Download links can be found via http://wissap.iiit.ac.in/blizzard2015/
* You will be sent a username and password to download the data after accepting the [http://wissap.iiit.ac.in/blizzard2015/license.html license]. Download links can be found via http://wissap.iiit.ac.in/blizzard2015
 
* Note that the official release for this challenge is only the speech data (wav/ directory), sampled at 16 kHz, and the corresponding text in UTF-8 format (train.done.data).  
* Note that the official release for this challenge is only the speech data (wav/ directory), sampled at 16 kHz, and the corresponding text in UTF-8 format (train.done.data).  


* Development tools, available via : http://www.cstr.ed.ac.uk/projects/blizzard/, include:
=== English (pilot task) ===
** Submitted synthetic speech and listener scores for some previous Blizzard Challenges, which may be helpful during development  
 
* About 2 hours of British English speech data in from a single female talker will be released by approx. 23rd February 2015. This is a subset of a larger dataset that will be used as the main task in the Blizzard Challenges 2016 and 2017.
 
* The speech comes from professional audiobooks produced by [http://www.usborne.com Usborne Publishing].
 
* A sentence-level alignment between text and speech is provided by [http://www.toshiba.eu/eu/Cambridge-Research-Laboratory/ Toshiba's Cambridge Research Laboratory].
 
* You will be sent a username and password to download the data after accepting the [http://www.cstr.ed.ac.uk/projects/blizzard/2015/usborne_blizzard2015/license.html license]. Download links can be found via http://www.cstr.ed.ac.uk/projects/blizzard/2015/usborne_blizzard2015
 
=== Development tools ===
 
* available via : http://www.cstr.ed.ac.uk/projects/blizzard/, which include submitted synthetic speech and listener scores for some previous Blizzard Challenges, which may be helpful during development
 
=== Questionnaire ===


* Questionnaire:
* Download the questionnaire (for the main IL tasks only), complete it, and return it by the deadline given in the Timeline. [[media:questionnaire.doc | Questionnaire]]
** Download the questionnaire, complete it, and return it by the deadline given in the Timeline. http://wissap.iiit.ac.in/blizzard2015/Blizzard2015_Ques.txt


== Mailing list ==
== Mailing list ==
Line 44: Line 72:
The timeline shown on this web page is the official one and supercedes those shown in announcements - it is subject to change, but we will try to follow it as closely as possible. Note that we will not consider any requests from participants to change the synthetic speech submission date or the paper submission date!
The timeline shown on this web page is the official one and supercedes those shown in announcements - it is subject to change, but we will try to follow it as closely as possible. Note that we will not consider any requests from participants to change the synthetic speech submission date or the paper submission date!


   Feb 15    2015  -  databases to be released
   Feb       2015  -  databases to be released
   Apr 21    2015  -  test sentences released to participants
   Apr 21    2015  -  test sentences released to participants
   Apr 28   2015  -  participants submit synthetic speech (by midnight PDT)
   Apr 30   2015  -  participants submit synthetic speech (by midnight PDT)
   May      2015  -  evaluation systems go live
   May      2015  -  evaluation systems go live
   Jun 30    2015  -  end of evaluation period
   Jun 30    2015  -  end of evaluation period
Line 52: Line 80:
   Aug  1    2015  -  deadline to submit workshop papers
   Aug  1    2015  -  deadline to submit workshop papers
   Aug 7    2015  -  notification of acceptance
   Aug 7    2015  -  notification of acceptance
   Sep 6    2015  -  [http://www.interspeech2015.org Interspeech 2015, Germany]
   Sep 6    2015  -  [http://www.interspeech2015.org Interspeech 2015, Dresden, Germany]
  Sep 11    2015 -  Blizzard Challenge workshop, Berlin, Germany


== Workshop ==
== Workshop ==

Latest revision as of 14:16, 5 December 2018

DeitY: Gov. of India, Apple Inc., and Google Inc., have generously provided financial support to the Blizzard Challenge 2015.

Google UK has generously provided financial support to the Blizzard Challenge 2015

This Blizzard Challenge has finished

  • Please do not try to register for this challenge - you are too late!
  • The data are no longer available via this site
  • However, related data may be available from https://www.iitm.ac.in/donlab/tts/index.php
  • The remainder of this page is left as a record of the Challenge


Read these first

Registration and license agreement

1. Register by emailing blizzard@festvox.org. We need to know your team name, the name of the main contact person, your affiliation, and contact details including email address, postal address and phone number. Please state which datasets you want to download and which tasks you are planning to submit entries to (this is non-binding, but is helpful for our planning).

2. Complete the license for Indian language data, which can be accessed via http://wissap.iiit.ac.in/blizzard2015/

3. Optionally, complete the license for the English pilot data, which can be accessed via http://www.cstr.ed.ac.uk/projects/blizzard/2015/usborne_blizzard2015

Data download

Indian Languages

  • About 4 hours of speech data is available in each of three Indian languages (Hindi, Tamil and Telugu), and about 2 hours of speech data in each of other three Indian languages (Marathi, Bengali and Malayalam), all recorded by native professional speakers in high quality studio environments. Text is provided in UTF-8 format. No other information, such as segment labels, is provided.
  • These speech databases are provided by the group of institutions: IIT-Madras, IIIT-Hyderabad, SSNCE, CDAC Trivandrum, CDAC Mumbai, CDAC Kolkata.
  • Note that the official release for this challenge is only the speech data (wav/ directory), sampled at 16 kHz, and the corresponding text in UTF-8 format (train.done.data).

English (pilot task)

  • About 2 hours of British English speech data in from a single female talker will be released by approx. 23rd February 2015. This is a subset of a larger dataset that will be used as the main task in the Blizzard Challenges 2016 and 2017.

Development tools

Questionnaire

  • Download the questionnaire (for the main IL tasks only), complete it, and return it by the deadline given in the Timeline. Questionnaire

Mailing list

There is a mailing list for discussion and announcements for the challenge:

 blizzard-discuss@festvox.org

Participants must join the list by sending a message to majordomo@festvox.org with the following line in the body of the message

 subscribe blizzard-discuss

Once you are a member you will be able to mail messages to blizzard-discuss@festvox.org

Timeline

The timeline shown on this web page is the official one and supercedes those shown in announcements - it is subject to change, but we will try to follow it as closely as possible. Note that we will not consider any requests from participants to change the synthetic speech submission date or the paper submission date!

  Feb       2015  -  databases to be released
  Apr 21    2015  -  test sentences released to participants
  Apr 30    2015  -  participants submit synthetic speech (by midnight PDT)
  May       2015  -  evaluation systems go live
  Jun 30    2015  -  end of evaluation period
  Jul 15    2015  -  release of results 
  Aug  1    2015  -  deadline to submit workshop papers
  Aug 7     2015  -  notification of acceptance
  Sep 6     2015  -  Interspeech 2015, Dresden, Germany
  Sep 11    2015 -  Blizzard Challenge workshop, Berlin, Germany

Workshop

Information on the workshop can be found here: Blizzard Challenge 2015 Workshop

Any questions?

  • Please contact blizzard@festvox.org if you have any questions

Previous challenges