Interesting approach if you are thinking to leverage BERT in Neural Machine Translation. arxiv.org/abs/2002.06823 MT Group at FBK

Phoneme-based semi-unsupervised reduction of speech data has been shown to be helpful for some speech tasks at #ICASSP2020. Will this become a standard approach for other tasks such as speech translation? cmsworkshops.com/ICASSP2020/Pap… MT Group at FBK

Our paper on "Contextualized Translation of Automatically Segmented Audio" has been accepted to #Interspeech2020. See you virtually there! @mdigangiPA MT Group at FBK

Working on speech translation? Tired of the same old target languages? 🥱 New version of MuST-C now available! 📌 Extended with 6 new directions: English➡️Arabic/Chinese/Czech/Persian/Turkish/Vietnamese ict.fbk.eu/must-c/ #nlproc #machinetranslation #speechtranslation

Another application of Reinforcement Learning in Machine Translation: arxiv.org/pdf/2010.03732… Lexical cohesion & coherence in document-level translation can be optimized by RL. by: @ijauregi, N. Esmaili, Reza Haffari & M. Piccardi MT Group at FBK

Biased gender representations can be a matter of stylistic nuance. Can they be automatically rewritten to ensure more agency and power? From my #EMNLP2020 highlights, "PowerTransformer: Unsupervised Controllable Revision for Biased Language Correction" aclweb.org/anthology/2020…

Our paper on "Breeding Gender-aware Direct Speech Translation Systems" will be presented on Thursday 10th at 15:30-16:00 CEST. BeatriceSavoldi MT Group at FBK Matteo Negri Luisa Bentivogli Marco Turchi

Presenting our #coling2020 paper aclweb.org/anthology/2020… The Two Shades of Dubbing in Neural Machine Translation at poster session16, Dec 10th @ 16:00 CET Alina Karakanta , Supratik Bhattacharya, Shravan Nayak, timobaumann , Matteo Negri , Marco Turchi MT Group at FBK

What are the challenges NLP tools facing to leverage user logs? Here is a comprehensive overview of Reinforcement Learning challenges in offline setting: arxiv.org/pdf/2011.02511… MT Group at FBK

Starting our #EAMT sponsored project: Towards a methodology for evaluating automatic subtitling We make the first steps towards a methodology for evaluating multiple aspects of quality in automatically generated subtitles. Stay tuned!🥳 MT Group at FBK EAMT

Come and join our team! We offer a fully funded 3-year PhD position: Flexibility and Robustness in Speech Translation iecs.unitn.it/education/admi… Full details for application: iecs.unitn.it/education/admi… Deadline: April 15, 2021 Università di Trento Fondazione Bruno Kessler - FBK #NLProc

Un asterisco sul genere - Consulenza Linguistica - Accademia della Crusca accademiadellacrusca.it/it/consulenza/…

Our student @at_amir ready to discuss his PhD research on machine translation for machines. Good luck for your viva! #nlproc #machinetranslation #phdviva Marco Turchi Matteo Negri

So much gratitude for everyone who was part of this journey!🙏 Special thanks to my supervisors Marco Turchi & Matteo Negri, and MT Group at FBK colleagues. #PhDone

🙋🏻‍♀️Soy Flor, investigadora predoctoral en Procesamiento del Lenguaje Natural #PLN #TLH👩🏻‍💻 en NLP_SINAI y acepto el interesante reto propuesto por Red Divulga, Crue Universidades y Universidad de Jaén para divulgar mi tesis doctoral en menos de 20 tweets. ¡Vamos a ello! Abro #HiloTesis 👇 🧵

📢 One week left to apply for a fully funded 3-year PhD position in our group: ⭐️ Application-oriented Speech Translation ⭐️ iecs.unitn.it/education/admi… Full details for application: iecs.unitn.it/education/admi… 📆 Deadline May 16, 2022 Università di Trento #NLProc Matteo Negri Marco Turchi

You have automatically generated subtitles and don't know how to evaluate their segmentation. Even worse, your text is not the same as your reference text. Sigma is the answer for you. Check out our new #LREC paper on evaluating #subtitle segmentation arxiv.org/abs/2205.09360

Our student Alina Alina Karakanta ready to present her work on automatic subtitling in her PhD viva #nlproc #phdviva

🚀📰 Thrilled to see today’s EL PAÍS Tecnología featuring my interview on #AI and #NLProc! I discuss my research on detecting hate speech, biases, and stereotypes, as well as the recent Young Researchers Award I received from Sociedad Científica Informática de España and Fundación BBVA. elpais.com/tecnologia/202…