O
Olli Viikki
Researcher at Nokia
Publications - 25
Citations - 645
Olli Viikki is an academic researcher from Nokia. The author has contributed to research in topics: Hidden Markov model & Speaker recognition. The author has an hindex of 9, co-authored 25 publications receiving 626 citations.
Papers
More filters
Journal ArticleDOI
Cepstral domain segmental feature vector normalization for noise robust speech recognition
Olli Viikki,Kari Laurila +1 more
TL;DR: A segmental feature vector normalization technique is proposed which makes an automatic speech recognition system more robust to environmental changes by normalizing the output of the signal-processing front-end to have similar segmental parameter statistics in all noise conditions.
Proceedings ArticleDOI
Multi-Lingual Speaker-Independent Voice User Interface For Mobile Devices
TL;DR: The system presented in this paper is the first of its kind to support both speech recognition and speech synthesis in more than 40 languages in embedded devices with strict memory and performance requirements.
Patent
Speech recognition from overlapping frequency bands with output data reduction
TL;DR: In this paper, a speech recognition feature extractor includes a time-to-frequency domain transformer for generating spectral values in the frequency domain from a speech signal; a partitioning means for generating a first set and an additional set of spectral values, where the first and the additional sets of values comprise at least one common spectral value.
Patent
Text-to-speech and midi ringing tone for communications devices
TL;DR: In this article, a method and device for providing a mixed audible signal which includes a ringing tone and the identity of a calling party upon receiving the incoming call of the calling party is presented.
Proceedings ArticleDOI
On a practical design of a low complexity speech recognition engine
TL;DR: The main design features of a low complexity speech recognition engine targeted for mobile devices are outlined and how these techniques can be successfully combined in order to achieve various design targets with minimized impact on the recognition performance is shown.