MLLP member Javier Iranzo Sánchez defended successfully his PhD thesis on “Streaming Neural Speech Translation” at Universitat Politècnica de València last Friday 29 September 2023.
Javier’s work for his dissertation has focused, quoting its abstract, on “researching and developing key techniques necessary for a successful streaming speech translation solution“. This entailed: “first, the new multilingual ST dataset Europarl-ST is collected for development and evaluation; […] then, a streaming-ready segmenter component is developed; […] this segmenter is then integrated and the components of the cascade are jointly optimized to achieve an appropriate quality-latency trade-off; […] new evaluation methods are proposed for streaming ST evaluation; […] lastly, a novel method is presented for improving translation quality through the use of contextual information”. Javier’s advisors were the MLLP’s Dr Alfons Juan Ciscar and Dr Jorge Civera Saiz.
The full text of the dissertation can be found at the UPV’s institutional repository RiuNet: https://doi.org/10.4995/Thesis/10251/199170
The examining committee for Javier’s PhD defence was made up of Dr Francesc Casacuberta Nolla (Univ. Politècnica de València), Dr Jesús Andrés Ferrer (Nuance Communications) and Dr Marco Turchi (Fondazione Bruno Kessler). They awarded Javier’s dissertation the qualification of Excellent.
After his successful PhD defence, Javier has earned the title “Doctor en Informàtica” (PhD in Computer Science) according to the Valencian and Spanish university systems.
Some of the most significant results covered in Javier’s PhD thesis have also been featured in his previous publications.
Everyone at the MLLP are proud to express congratulations to Javier on his hard-earned new title!
