Ulpu Remes

Doctoral student, M.Sc. (Tech.)

About Me

A doctoral student at the Aalto University School of Science, working on noise-robust speech recognition in the Speech Group. Visited the Tokuda & Lee Laboratory in Nagoya Institute of Technology on the academic year 2010-2011.

Publications

• Hannu Pulakka, Ulpu Remes, Santeri Yrttiaho, Kalle J. Palomäki, Mikko Kurimo, and Paavo Alku (to appear). Bandwidth extension of telephone speech to low frequencies using sinusoidal synthesis and a Gaussian mixture model. IEEE Trans. Audio, Speech, and Language Processing, to appear. [link]
• Ulpu Remes, Kalle J. Palomäki, Tapani Raiko, Antti Honkela, and Mikko Kurimo (2011). Missing-feature reconstruction with a bounded nonlinear state-space model. IEEE Signal Processing Letters, vol 18, issue 10, pp. 563-566, October 2011. [link]
• Heikki Kallasjoki, Sami Keronen, Guy J. Brown, Jort F. Gemmeke, Ulpu Remes, and Kalle J. Palomäki (2011). Mask estimation and sparse imputation for missing data speech recognition in multisource reverberant environments. Proc. CHIME, pp. 58-63, Florence, Italy, September 2011.
• Heikki Kallasjoki, Ulpu Remes, Jort F. Gemmeke, Tuomas Virtanen, and Kalle J. Palomäki (2011). Uncertainty measures for improving exemplar-based source separation. Proc. INTERSPEECH, pp. 469-472, Florence, Italy, August 2011. [link]
• Ulpu Remes, Yoshihiko Nankaku, and Keiichi Tokuda (2011). GMM-based missing feature reconstruction on multi-frame windows. Proc. INTERSPEECH, pp. 1665-1668, Florence, Italy, August 2011. [link]
• Hannu Pulakka, Ulpu Remes, Santeri Yrttiaho, Kalle J. Palomäki, Mikko Kurimo, and Paavo Alku (2011). Low-frequency bandwidth extension of telephone speech using sinusoidal synthesis and Gaussian mixture model. Proc. INTERSPEECH, pp. 1181-1184, Florence, Italy, August 2011. [link]
• Hannu Pulakka, Ulpu Remes, Kalle Palomäki, Mikko Kurimo, and Paavo Alku (2011). Speech bandwidth extension using Gaussian mixture model based estimation of the highband mel spectrum. Proc. ICASSP, pp. 5100-5103, Prague, Czech Republic, May 2011. [link]
• Jort F. Gemmeke, Bert Cranen, and Ulpu Remes (2011). Sparse imputation for large vocabulary noise robust ASR. Computer Speech & Language, vol 25, issue 2, pp. 462-479, April 2011. [link]
• Jort F. Gemmeke, Ulpu Remes, and Kalle J. Palomäki (2010). Observation uncertainty measures for sparse imputation. Proc. INTERSPEECH, Makuhari, Japan, pp. 2262-2265, September 2010. [link]
• Sami Keronen, Ulpu Remes, Kalle J. Palomäki, Tuomas Virtanen, and Mikko Kurimo (2010). Comparison of noise robust methods in large vocabulary speech recognition. Proc. EUSIPCO, pp. 1973-1977, Aalborg, Denmark, August 2010. [pdf]
• Ulpu Remes, Kalle J. Palomäki, and Mikko Kurimo (2009). Robust automatic speech recognition using acoustic model adaptation prior to missing feature reconstruction. Proc. EUSIPCO, pp. 535-539, Glasgow, Scotland, UK, August 2009. [pdf]
• Ulpu Remes, Kalle J. Palomäki, and Mikko Kurimo (2008). Missing feature reconstruction and acoustic model adaptation combined for large vocabulary continuous speech recognition. Proc. EUSIPCO, Lausanne, Switzerland, August 2008. [pdf]
• Ulpu Remes, Kalle J. Palomäki, and Mikko Kurimo (2008). Speaker adaptation combined with missing data reconstruction (A). J. Acoust. Soc. Am. vol 123, issue 5, p. 3181, May 2008. [link]
• Ulpu Remes, Janne Pylkkönen, and Mikko Kurimo (2007). Segregation of speakers for speaker adaptation in TV news audio. Proc. ICASSP, pp. 481-484, Honolulu, Hawaii, USA, April 2007. [link]

Other

• Kalle J. Palomäki, Ulpu Remes, and Mikko Kurimo, Eds. (2009). Studies on Noise Robust Automatic Speech Recognition. Technical report TKK-ICS-R19, Helsinki University of Technology, 2009. [pdf]
• Master's thesis: Speaker-Based Segmentation and Adaptation in Automatic Speech Recognition. Helsinki University of Technology, 2007. [pdf]