Research in the Speech group is funded by several collaborative research projects in Academy of Finland, Tekes, European Commission and various foundations and by individual grants. The main research projects where we are currently involved in are:
COIN (2012 - 2017). The centre of excellence in computational inference funded by the Academy of Finland. The research is in collaboration with the computer science department at Aalto and University of Helsinki.
SIAK (2014 - 2017). "Say it again, kid!" is a research project funded by the Academy of Finland that applies automatic speech recognition and rating for games that practise second language pronunciation learning. The research is in collaboration with the brain research unit at University of Helsinki.
TELLme (2016 - 2018). "Technology enhanced language learning for mobile environments" is a research project that won the Tekes Challenge Finland funding. The research is in collaboration with the brain research unit at University of Helsinki and several companies.
DigiTala (2015 - 2017). A project where we study assessing oral second language skills in high-
stakes examinations. The research is in collaboration with the Finnish matriculation examination board and teacher education and phonetics at University of Helsinki.
Respeak (2016 - 2018). A project where we study automatical speech-to-text transformation for subtitling broadcasts and other audiovisual events for the hard of hearing and other viewers who have problems in hearing the audio.
MeMAD (2018 - 2020). "Methods for Managing Audiovisual Data: Combining Automatic Efficiency with Human Accuracy" is a European Commission Horizon 2020 ICT-20-2017-RIA project that is coodinated by us. In the project we will develop new generation of automatic methods to describe speech, audio and visual content in written language for indexing, browsing, translating, viewing and listening. The other partners are University of Helsinki, Surrey, Eurecom and several companies.
Long term projects supported by various organizations
- Language modeling: Mikko Kurimo
- Speaker adaptation: Mikko Kurimo
- Noise robust speech recognition: Kalle Palomäki
- Morpho Challenge: Morpho Challenge
Finished projects (since 2000)
- Finnish Radio and Television Speech Database (Finnish Academy): Mikko Kurimo
- SA-PUHE (Finnish Academy): Mikko Kurimo
- USIX-STT (TEKES): Mikko Kurimo
- FENIX-PUMS (TEKES): Mikko Kurimo
- SEPEMCO: Mikko Kurimo, Kalle Palomäki
- EMIME (2008-2011) Effective Multilingual Interaction in Mobile Environments (EMIME) is a research project funded by EU FP7, in collaboration with University of Edinburgh, University of Cambridge, Nagoya Insitute of Technology, Idiap, and Nokia.
- New automatic annotation and indexing methods for large volumes of audiovisual data (2012-2014). A research project funded by the Academy of Finland in the call for supporting research in the Strategic centres for Science, Technology and Innovation.
- Simple4All (2011-2014). The SIMPLE4ALL project will create speech synthesis technology that learns from data with little or no expert supervision and continually improves itself, simply by being used. It is a research project funded by EU FP7, in collaboration with University of Edinburgh, University of Helsinki, Technical University of Cluj-Napoca, and Technical University of Madrid.
- Mobster (2011-2013). Mobile and integrated dictation and communication application for health care, Mobster, is a large nation-wide research project funded by Tekes. Our academic partners are at University of Tampere, University of Helsinki and University of Turku.
- Academy Research Fellowship in Noiserobust ASR (2010-2015). "Noise Robust Automatic Speech Recognition: Searching for Answers from Human Hearing" is a five year research fellowship of Kalle Palomäki funded by the Academy of Finland. The fellowship is supplemented by a grant for establishment of a research team in noise robust ASR.
- CSI-Speech (2010-2014). Computational Speech Inversion (CSI-Speech) is a research project funded by the Academy of Finland in the research programme LASTU.
- NextMedia (2010-2013). Speech is part of WP4 coordinated by the Department of Media Technology at Aalto UniversitySchool of Science which is a project funded by ICT SHOK Next Media. The collaborators in speech processing are Lingsoft and VTT.
- UI-ART (2008-2013). Urban contextual information interfaces with multimodal augmented reality (UI-ART) is a research project funded by TKK under the Multidisciplinary Institute for Digitalisation and Energy (MIDE) research programme.
- PinView (2008-2011) Personal Information Navigator Adapting Through Viewing (PinView) is a research project funded by EU FP7, in collaboration with University of Southampton, University College London, University of Leoben, Xerox Research Centre Europe, and celum gmbh.
- MMR (2008-2011) Mobile Mixed Reality (MMR) is a research project funded by ICT SHOK Devices and Interoperability Ecosystem research programme. The collaborators are the Department of Media Technology at Aalto University School of Science, Nokia Research Center, and VTT.
- Perso (2011-2012) Personalized Hidden Markov Modeling (Perso) based Text-to-Speech Synthesis: Assistive Technology for People with Communication Disabilities is a research project funded by Tekes. Our academic partners are Department of Signal Processing and Acoustics at Aalto University and Department of Speech Sciences at Helsinki University.