We have open Postdoc and Master's student positions (filled in 11.1.2013).
Sami Virpioja will defend his dissertation "Learning Constructions of Natural Language: Statistical Models and Evaluations". Press Release. The opponents will be Prof. Brian Roark from Oregon Health & Science University and Dr. Krister Linden from University of Helsinki.
Mikko Kurimo receives a professorship in Speech and Language Processing at Department of Signal Processing and Acoustics, Aalto University School of Electrical Engineering. In 2013 the speech recognition research group will then move from CS to the EE building to emphasize the department's focus in speech and language processing.
At Interspeech 2012 Janne Pylkkönen presents: "Optimization-based control for the extended baum-welch algorithm" and "Improving discriminative training for robust acoustic models in large vocabulary continuous speech recognition", and Kalle Palomäki: "Perceptual compensation for the effects of reverberation on consonant identification: A comparison of human and machine performance".
Ville Turunen defended successfully his dissertation "Morph-Based Speech Retrieval: Indexing Methods and Evaluations of Unsupervised Morphological Analysis". Press Release. The opponent was Prof. Murat Saraclar from Bogazici University, Istanbul who is currently visiting Google.
New group picture taken: click here.
Speech synthesis demo samples added to the demo page.
We have an open project in the department's summer internship program. The DL for applications is February 6.
The first meeting of the speech group ex-members and current members took place in Otaniemi. We heard how the work has been like in start-ups and larger companies, such as: Rosetta Stone, Lingsoft, Alpha Sense, Nokia Siemens Networks, Nokia Research Center, and Process Vision.
Several new projects are starting, follow funded projects for more info.
We have an open position for preparing a Master's thesis in a research project (filled in 16.1.2012).
Speaker adaptation for speech synthesis was participating in the Pispala play at Radioteatteri in Finnish radio station YLE RADIO 1.
Our joint paper with the Aalto Acoustics laboratory: "Low-Frequency Bandwidth Extension of Telephone Speech Using Sinusoidal Synthesis and Gaussian Mixture Model" by Hannu Pulakka, Ulpu Remes, Santeri Yrttiaho, Kalle Palomäki, Mikko Kurimo, and Paavo Alku (presented by Hannu Pulakka) wins the ISCA Award for the best student paper of Interspeech 2011.
At Interspeech 2011 Sami Keronen presents: "Noise Robust Feature Extraction Based on Extended Weighted Linear Prediction in LVCSR", Heikki Kallasjoki: "Uncertainty measures for improving exemplar-based source separation", Ulpu Remes: "GMM-based missing-feature reconstruction on multi-frame windows", Reima Karhila: "Rapid adaptation of foreign-accented HMM-based Speech Synthesis", and Rama Sanand Doddipatla: "A Study on Combining VTLN and SAT to Improve the Performance of Automatic Speech Recognition".
We have an open postdoc position (already filled).
The Final Review of the FP7 project EMIME (Effective Multilingual Interaction in Mobile Environments, 2008-2011) is successfully terminated achieving the highest possible grade: "Good to excellent progress (the project has fully achieved its objectives and technical goals for the period and has even exceeded expectations)."
Peter Smit presents: "Using stacked transformations for recognizing foreign accented speech" and Reima Karhila: "Speaker similarity evaluation of foreign-accented speech synthesis using HMM-based speaker adaptation" at IEEE International Conference on Acoustics, Speech and Signal Processing in Prague, Czech Republic.
Our speech recognition system was demonstrated in the news of Russian national TV channel Vesti. The same news flash presents also the intelligent information retrieval system of the UI-ART/MIDE project using a near-eye display and gaze tracker developed by the Nokia Research Center.
Several new projects are starting, follow funded projects for more info.
We have an open position in a Master's thesis project (filled in 21.1.2011).
Our homepage has a new appearance.
Our annual course on speech recognition receives the First Prize of small courses (less than 50 participants) in Computer Science and Engineering based on student feedback in 2009-2010.
Reima Karhila presents: "Unsupervised cross-lingual speaker adaptation for accented speech recognition" at IEEE Workshop on Spoken Language Technology in Berkeley, California.
The speaker adaptation results by the "researchers at Aalto University" appear in the main news of the Finnish TV channel Nelonen
Our goal is to generally improve the speech recognition methodology with the help of the new algorithms developed in our department. Speech recognition offers challenging benchmarking tasks for efficient algorithms that can process and learn to represent large quantities of data. In addition to improving the acoustic models of phonemes we aim at developing new learning statistical language models for difficult large vocabulary continuous speech recognition tasks. The relevant pilot applications in our group range from unlimited vocabulary continuous dictation in different languages to audio indexing, adaptive speech synthesis, and speech-to-speech translation.
At the national Center of Excellence in Computational Inference Research the research group belongs to the flagship of Intelligent Information Access. Head of the group is Mikko Kurimo. In language modeling, speech translation and video indexing we work closely together with the Adaptive Natural Language Processing, Computational Cognitive Systems, Content-Based Image and Information Retrieval research groups.
| Mikko Kurimo | Kalle Palomäki | Ulpu Remes | Heikki Kallasjoki | Reima Karhila |
| Teemu Ruokolainen | Andre Mansikkaniemi | Sami Keronen | Peter Smit | Seppo Enarvi |
| Matti Varjokallio |