- SOM based density function approximation for mixture density HMMs
- LATENT SEMANTIC INDEXING BY SELFORGANIZING MAP Mikko Kurimo and Chafic Mokbel
- 1 Using SOM and LVQ for HMM training Mikko Kurimo
- Unlimited Vocabulary Speech Recognition Based on Morphs Discovered in an Unsupervised Manner
- To recover from speech recognition errors in spoken document retrieval Mikko Kurimo and Ville Turunen
- Vocabulary Decomposition for Estonian Open Vocabulary Speech Recognition
- 1 Acoustic modeling The general goal of automatic speech recognition (ASR) is to understand normal human
- Ma 87 UDC 681.327.12:534.75 POLYTECHNICA
- In Proceedings of ICASSP, volume 1, pages 639--642, New York, [3] BiingHwang Juang. Maximum likelihood estimation for mixture
- [5] Teuvo Kohonen. The SelfOrganizing Map. In Proceedings of the IEEE, pages 1464--1480, 1990. [6] Teuvo Kohonen, Jari Kangas, Jorma Laaksonen, and Kari Torkkola. LVQPAK: A program package for the correct
- CORRECTIVE TUNING BY APPLYING LVQ FOR CONTINUOUS DENSITY AND SEMICONTINUOUS
- Hybrid training method for tied mixture density hidden Markov models using
- Published in the proceedings of ICANN91, Espoo, Finland, June 2428, 1991, pp. 771776 STATUS REPORT OF THE FINNISH
- ffl Tuning: 1 cycle with all training samples decreasing the learning rate down from 0.02 and using learning
- USING THE SELFORGANIZING MAP TO SPEED UP THE PROBABILITY DENSITY ESTIMATION FOR SPEECH RECOGNITION WITH MIXTURE
- SEGREGATION OF SPEAKERS FOR SPEAKER ADAPTATION IN TV NEWS AUDIO Ulpu Remes, Janne Pylkkonen, Mikko Kurimo
- Indexing Confusion Networks for Morph-based Spoken Document Retrieval
- Using Phone Durations in Finnish Large Vocabulary Continuous Speech Recognition
- Duration Modeling Techniques for Continuous Speech Recognition Janne Pylkkonen and Mikko Kurimo
- Training Mixture Density HMMs with SOM and LVQ
- Comparison of Subspace Methods for Gaussian Mixture Models in Speech Recognition
- Using Latent Semantic Indexing for Morph-based Spoken Document Retrieval Ville T. Turunen, Mikko Kurimo
- Compact n-gram models by incremental growing and clustering of histories Sami Virpioja and Mikko Kurimo
- Unlimited vocabulary speech recognition for agglutinative languages Mikko Kurimo1
- Retrieving speech correctly despite the recognition errors
- LANGUAGE MODELING STRUCTURES IN AUDIO TRANSCRIPTION FOR RETRIEVAL OF HISTORICAL SPEECHES
- HELSINKI UNIVERSITY OF TECHNOLOGY Faculty of Information Technology
- Seattle, Washington, USA IMPROVING VOCABULARY INDEPENDENT HMM DECODING RESULTS BY USING
- Large Vocabulary Statistical Language Modeling for Continuous Speech Recognition in Finnish
- MORPHOLOGICALLY MOTIVATED LANGUAGE MODELS IN SPEECH RECOGNITION
- 1 Speech Recognition Mikko Kurimo, Panu Somervuo
- 1 Speech Recognition Mikko Kurimo, Panu Somervuo
- Using LVQ and SOM in speech recognition with multiple feature streams M. Kurimo and P. Somervuo
- Decoder Issues in Unlimited Finnish Speech Recognition Teemu Hirsimki and Mikko Kurimo
- SEGMENTAL LVQ3 TRAINING FOR PHONEMEWISE TIED MIXTURE DENSITY HMMS
- INDEXING SPOKEN AUDIO BY LSA AND SOMS Mikko Kurimo
- Indexing Audio Documents by using Latent Semantic Analysis and SOM Mikko Kurimo a
- Thematic Indexing of Spoken Documents by Using SelfOrganizing Maps
- vergence to low error rates occurs faster and yields (in the av erage) better models. If the mean vectors of the multivariate
- FAST LATENT SEMANTIC INDEXING OF SPOKEN DOCUMENTS BY USING SELF-ORGANIZING MAPS
- An Evaluation of a Spoken Document Retrieval Baseline System in Finnish Mikko Kurimo and Ville Turunen
- Combinations of adaptive vector quantization met-hods and continuous density hidden Markov models
- Using LVQ to enhance continuous density and semicontinuous hidden Markov models for phonemes
- On Lexicon Creation for Turkish LVCSR Kadri Hacioglu, Bryan Pellom
- LANGUAGE MODEL ADAPTATION IN SPEECH RECOGNITION USING DOCUMENT MAPS
- Methods for Combining Language Models in Speech Recognition Simo Broman and Mikko Kurimo
- An Eciently Focusing Large Vocabulary Language Model
- Bruges, Belgium Self Organization in Mixture Densities of
- INDEPENDENT Mikko.Kurimo@hut.
- Using LVQ and SOM to train mixture density HMMs for phonemes Mikko Kurimo
- 1 Using SOM and LVQ for HMM training Mikko Kurimo
- LVQ and SOMs in training continuous and semicontinuous density hidden Markov models in phonemic decoding
