Deep neural networks for oral and written language processing
PhD Student: Antoine Caubrière
Advisor(s): Yannick Estève (LIUM, LST)
Co-advisor(s): Antoine Laurent (LIUM, LST) & Emmanuel Morin (LS2N)
Funding: RAPACE Project
The aim of this thesis is to develop a named entity recognition system in an audio stream that will rely solely on a deep neural network. Until now, this task has been carried out via successive processing chains. Also, speech and named entity recognitions fully neuronal tasks have been greatly improved in recent years.
It is a question of exploiting these advances and extending them to design end-to-end systems that are fully optimized for the named entity recognition task directly from audio, without explicitly taking into account the performance obtained on intermediate levels.