Spoken Language Systems Lab

From HLT@INESC-ID

Revision as of 15:48, 21 November 2014 by Rdmr (talk | contribs)
L²F - INESC ID Lisboa
Events
News

INESC ID's Spoken Language Systems Laboratory (L²F - Laboratório de sistemas de Língua Falada) was created in January 2001, bringing together several research groups which could potentially add relevant contributions to the area of computational processing of spoken language for Portuguese. The lab is integrated in the research unit of Interactive Intelligent Systems.

<php> print 'Gopala`s PhD thesis work
'; print '<iframe width="300" height="200" src="http://www.youtube.com/embed/LzYn6kdcIIc" frameborder="0" allowfullscreen></iframe>'; print '
'; print 'ASR + SMT + TTS/SPAM + INTENT TRANSFER'; </php>

PT original (ASR+SMT subtitles)

ASR + SMT + TTS (no SPAM)

ASR + SMT + TTS/SPAM

More demos from L2F on YouTube: S2S translation VITHEA 3D office PT-STAR


Research Opportunities - The Spoken Language Systems Lab is looking for students interested in pursuing a PhD in Language Technologies. Post-Doc positions are also open. The typical PhD program in our lab is held in cooperation with international Universities.



New Master Theses Themes at L2F (2013/14)



Next events:

Past events:

July 22th-29th, 2014
The 4th Lisbon Machine Learning School (LxMLS 2014) took place at Instituto Superior Técnico. It was organized jointly by IST, the Instituto de Telecomunicações and the Spoken Language Systems Lab (L2F) of INESC-ID.

June 30th, 2014 (14h30; Amphitheatre Abreu Faro, IST Alameda)
Prof. Shrikanth Narayanan from the University of Southern California, USA, presented "Behavioral Signal Processing: Enabling human-centered behavioural informatics"

May 26th, 2014 (QA1.2, IST Alameda)
Prof. Chin-Hui Lee from the School of Electrical and Computer Engineering, Georgia Institute of Technology, presented "A Deep Neural Network Approach to Speech Enhancement".

May 2nd, 2014 (CMU)
Luís Marujo presented his PhD proposal, entitled "Temporal Contextualization of News Articles and Broadcast News".

Apr 16th, 2014 (CMU)
João Miranda presented his PhD proposal, entitled "Combining Multiple Parallel Streams for Improved Speech Processing".

Dec 3rd, 2013 (16h, room 336)
Joakim Gustafson presented Towards face-to-face conversations with social robots.

Oct 31st, 2013 (13h, room 020)
Bhiksha Raj presented Unsupervised semantic structure discovery for audio.


Participation in Conferences:

Oct 16-17
Helena Moniz and Vera Cabarrão attended the Pragmatic Markers, Discourse Markers and Modal Particles: What do we know and where do we go from here? Workshop, in Como, Italy.

Oct 2-4, 2014
Helena Moniz attended Capri Prominence Workshop, in Capri, Italy.

Sept 14-18
Ramon Fernandez Astudillo and Isabel Trancoso attended Interspeech 2014, the 15ht Annual Conference of the International Speech Communication Association, in Singapore.

Jul 6-11, 2014
Luís Marujo attended SIGIR 2014, the 37th annual international ACM SIGIR conference, in Queensland, Australia.

Jul 6-11, 2014
Joao Paulo Carvalho attended FUZZ-IEEE 2014, the 2014 IEEE International Conference on Fuzzy Systems, in Beijing, China.

May 26-31, 2014
Anabela Barreiro, Fernando Batista, Hugo Meinedo, Ângela Costa, Wang Ling and Rui Correia attended LREC 2014, the 9th edition of the Language Resources and Evaluation Conference, in Reykjavik, Iceland.




L2F researchers received the best paper award in the PROPOR 2014 conference with the paper "Improving Speech Recognition through Automatic Selection of Age Group – Specific Acoustic Models", authored by Annika Hämäläinen, Hugo Meinedo, Michael Tjalve, Thomas Pellegrini, Isabel Trancoso, and Miguel Sales Dias.

Vera Cabarrão M.A. thesis, entitled Respostas afirmativas em diálogos espontâneos em Português Europeu (Interface Prosódia-Sintaxe-Discurso), Faculty of Letters, University of Lisbon, received the Research Prize of the Portuguese Linguistics Society (APL) on 2014.

Project VITHEA (Virtual Therapist for Aphasia treatment) is part of an exhibition on mental health that will be inaugurated October 30th, 2014, at the Pavilhão do Conhecimento. For more information see [1].

Isabel Trancoso was appointed as ISCA Fellow for "her research in speech coding and significant and sustained contributions to speech science and technology for the Portuguese language".

Alberto Abad Gareta and Anna Pompili have appeared on the news again for their work on VITHEA (Virtual Therapist for Aphasia treatment).

Scientific Award - Fernando Batista was one the winners of the 2013 edition of the ISCTE-IUL scientific awards with the work "Bilingual Experiments on Automatic Recovery of Capitalization and Punctuation of Automatic Speech Transcripts" published in the IEEE Transactions on Audio, Speech, and Language Processing.

Alberto Abad Gareta and Anna Pompili have received a prize for their work on project VITHEA (Virtual Therapist for Aphasia treatment).

IEEE Fellow - Prof. Isabel Trancoso The Institute of Electrical and Electronics Engineers, IEEE, appointed Prof. Isabel Trancoso as a Fellow of this prestigious institution.




Voice Interaction


VoiceInteraction S.A. is a technology based company born in April 2008 as a spin-off from INESC-ID Lisboa. Its primary goal is to develop speech processing technologies. VoiceInteraction team is composed by researchers coming from INESC-ID Lisboa Spoken Language Systems Lab (L2F/INESC-ID).