1 code implementation • EACL 2021 • Ebrahim Ansari, Ond{\v{r}}ej Bojar, Barry Haddow, Mohammad Mahmoudi
SLTev reports the quality, latency, and stability of an SLT candidate output based on the time-stamped transcript and reference translation into a target language.
no code implementations • RANLP 2019 • Ebrahim Ansari, Zden{\v{e}}k {\v{Z}}abokrtsk{\'y}, Mohammad Mahmoudi, Hamid Haghdoost, Jon{\'a}{\v{s}} Vidra
In the experimental phase, using the hand-annotated Persian lexicon and two smaller similar lexicons for Czech and Finnish languages, we evaluated the effect of the training data size, different hyper-parameters settings as well as different RNN-based models.
no code implementations • EACL 2021 • Ond{\v{r}}ej Bojar, Dominik Mach{\'a}{\v{c}}ek, Sangeet Sagar, Otakar Smr{\v{z}}, Jon{\'a}{\v{s}} Kratochv{\'\i}l, Peter Pol{\'a}k, Ebrahim Ansari, Mohammad Mahmoudi, Rishu Kumar, Dario Franceschini, Chiara Canton, Ivan Simonini, Thai-Son Nguyen, Felix Schneider, Sebastian St{\"u}ker, Alex Waibel, Barry Haddow, Rico Sennrich, Philip Williams
This paper presents an automatic speech translation system aimed at live subtitling of conference presentations.
no code implementations • 24 Jun 2022 • Mohammad Mahmoudi, Nasser Sadati
Then, we introduce a novel meta-reinforcement learning framework to adapt the latent space of CGAN to the system's uncertainties as an optimal direct adaptive controller without any system identifier.