Short news

Meeting recognition project MINT started

In October 2014, we have started a TACR (Technology Agency of the Czech Republic) supported project "Meeting assistant - MINT". The goal is to build on the experience gathered in EU projects M4, AMI and AMIDA and come up with real working solutions for real business users.

The partners of the project are

  • Phonexia - responsible for integration and core speech mining SW
  • Lingea - multi-linguality and translation
  • TOVEK - summarization, hot-spot detection and other high levels of analysis

ASRU is over

Last week, two-year organizing efforts of BUT Speech@FIT culminated: ASRU took place in NH Hotel Olomouc. The conference run smoothly, the keynote and invited talks were very interesting and the continuous poster session was appreciated by most participants. Social program was successful as well, including the banquet with a folklore-rock music band and visits to the historical center of Olomouc. Some attendees also learned what "non-stop bar" means in Czech.

Conference website:

Mireia about to leave

Mireia Diez Sánchez from the University of the Basque Country (Spain) visited the group for three months to work on speaker recognition. She did her final presentation on Tuesday Decemer 17th and is heading back home to join her family before Christmas. Wishing her good journey back and looking forward to further cooperation!

Harish in Brno to help with FDLP

Harish Mallidi, a Ph.D. student of Prof. Hynek Hermnasky at Johns Hopkins University, came to Brno for 2 moths to help with integration of FDLP (frequency domain linear prediction) and other advanced feature extraction techniques investigated at JHU, with the speaker and speech recognition systems developed at BUT. Harish will also help the BUT team with forthcoming RATS evaluations that will take place in January.

Award of Czech ICT alliance for the best academia-industry cooperation in the domain of IT

On October 22, 2013, Phonexia and BUT Speech@FIT obtained the Award of Czech ICT alliance for the best academia-industry cooperation in the domain of IT. The official news is available here.

Kaldi workshop again in Brno

We are happy that Kaldi workshop has started at BUT in Brno last week. There are 16 people participating and working hard for the 9 days. Among others our guests are:

  • Dan Povey, Sanjeev Khudanpur, Gaurav Kumar and Jan Trmal from JHU, USA
  • Korbinian Riedhammer from ICSI, USA
  • Petr Motlicek and David Imseng from IDIAP, Switzerland
  • Thang Vu from KIT, Germany
  • and Filip Jurčíček and Ondřej Plátek from CUNI, Prague

World’s top scientists in speaker recognition together again – this year at Johns Hopkins

The world’s leading researchers in speaker recognition just can’t imagine spending summer without each other. After the success of Brno Speaker Recognition Workshop BOSARIS 2010 and BOSARIS 2012 (well, the 2012 took place a bit later than in summer), the group gathered again – this year hosted by the Center for Language and Speech Processing at the Johns Hopkins University.

ASRU receives an ONR support of $10,000

BUT received a grant of $10,000 from U.S. Office of Naval Research Global’s (ONRG) as a support of ASRU, in Olomouc, Czech Republic, from December 8-12 2013. The ASRU workshop meets every two years and has a tradition of bringing together top researchers from academia and industry in an intimate and collegial setting to discuss problems of our common interest. 2013 edition of ASRU is organized by BUT Speech@FIT group.

The grant will serve mainly to bring quality keynote and invited speakers to the conference, to make it affordable for students, and to have ASRU talks recognized and indexed at

We are grateful to Dr. Charles J. Holland, the Technical Lead of ONRG Prague Office, for support and help. Charlie is actually a "speech guy" and before his ONRG appointment, he served as Head of Unit at DARPA, overseeing several high-performance computing, speech recognition and machine translation programs.

Screaming competition 2013

This year the established screaming competition was held at our campus on April 4th on occasion of the closed-door day event. We want to thank all participants for their excellent and loud performance! The winner became Zuzana with her voice reaching earsplitting 130,5 dB. Congratulations!