Product Details
SW2 Robustní diarizace
Created: 2022
English title
SW2 Robust diarization
Type
software
License
not public
Authors
Karafiát Martin, Ing., Ph.D. (DCGM FIT BUT)
Diez Sánchez Mireia, M.Sc., Ph.D. (DCGM FIT BUT)
Švec Jan, Ing., Ph.D. (WBU in Pilsen)
Černocký Jan, prof. Dr. Ing. (DCGM FIT BUT)
Szőke Igor, Ing., Ph.D. (DCGM FIT BUT)
Šmídl Luboš, Ing., Ph.D. (WBU in Pilsen)
Zajíc Zbyněk, Ing., Ph.D. (WBU in Pilsen)
Diez Sánchez Mireia, M.Sc., Ph.D. (DCGM FIT BUT)
Švec Jan, Ing., Ph.D. (WBU in Pilsen)
Černocký Jan, prof. Dr. Ing. (DCGM FIT BUT)
Szőke Igor, Ing., Ph.D. (DCGM FIT BUT)
Šmídl Luboš, Ing., Ph.D. (WBU in Pilsen)
Zajíc Zbyněk, Ing., Ph.D. (WBU in Pilsen)
Keywords
robust diarization
Description
A system for diarization (determining who speaks when in a recording) was developed based on current work on a Bayesian approach using NN embeddings. The result is in the form of an application executable from the command line on standard Linux/Windows distribution. The documentation includes a description of the technical solution and implementation. The entire set of tools is packaged in a Docker container solution for high deployment flexibility.
Projects
Research groups
Speech Data Mining Research Group BUT Speech@FIT (VZ SPEECH)
Departments
Department of Computer Graphics and Multimedia FIT BUT (DCGM FIT BUT)
University of West Bohemia in Pilsen (WBU in Pilsen)
University of West Bohemia in Pilsen (WBU in Pilsen)