Online Speaker Diarization
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F14%3A43923808" target="_blank" >RIV/49777513:23520/14:43923808 - isvavai.cz</a>
Výsledek na webu
<a href="http://hdl.handle.net/11025/21259" target="_blank" >http://hdl.handle.net/11025/21259</a>
DOI - Digital Object Identifier
—
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
Online Speaker Diarization
Popis výsledku v původním jazyce
In automatic speech processing, speaker diarization is the task of distinguishing between different speakers within an audio recording and identifying the intervals in which they are active, or in other words, determining “Who spoke when?”. This is generally done without any prior knowledge about the actual identities and number of speakers. The information obtained from speaker diarization can be used in several areas, such as audio indexing and searching or for improving the performance of speech recognition systems. Some of these areas require the diarization to be done online. This represents a more difficult variant of the task and generally leads to a worsened performance compared to offline systems. An online diarization system was created based on the one proposed by Markov and Nakamura (2007). The goal was to further improve its performance.
Název v anglickém jazyce
Online Speaker Diarization
Popis výsledku anglicky
In automatic speech processing, speaker diarization is the task of distinguishing between different speakers within an audio recording and identifying the intervals in which they are active, or in other words, determining “Who spoke when?”. This is generally done without any prior knowledge about the actual identities and number of speakers. The information obtained from speaker diarization can be used in several areas, such as audio indexing and searching or for improving the performance of speech recognition systems. Some of these areas require the diarization to be done online. This represents a more difficult variant of the task and generally leads to a worsened performance compared to offline systems. An online diarization system was created based on the one proposed by Markov and Nakamura (2007). The goal was to further improve its performance.
Klasifikace
Druh
O - Ostatní výsledky
CEP obor
—
OECD FORD obor
20205 - Automation and control systems
Návaznosti výsledku
Projekt
—
Návaznosti
S - Specificky vyzkum na vysokych skolach
Ostatní
Rok uplatnění
2014
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů