Tomi Kinnunen, Evgeny Karpov, Pasi Fränti
Abstract — In speaker identification, most of the computation originates from the distance or likelihood computations between the feature vectors of the unknown speaker and the models in the...
Contents lists available at ScienceDirect Pattern Recognition Letters (2009)
Ville Hautamäki, Tomi Kinnunen, Pasi Fränti
journal homepage: www.elsevier.com/locate/patrec
APPLYING MFCC-BASED AUTOMATIC SPEAKER RECOGNITION TO GSM AND FORENSIC DATA (2009)
Tuija Niemi-laitinen, Juhani Saastamoinen, Tomi Kinnunen, Pasi Fränti
Speaker Profiler computer program for automatic speaker recognition has been developed in a research project funded by the Finnish Technology Agency. A vector quantization (VQ) matching approach is...
Voice Activity Detection Using MFCC Features and Support Vector Machine (2009)
Tomi Kinnunen, Evgenia Chernenko, Marko Tuononen, Pasi Fränti, Haizhou Li
We define voice activity detection (VAD) as a binary classification problem and solve it using the support vector machine (SVM). Challenges in SVM-based approach include selection of representative...
On the Use of Long-Term Average Spectrum in Automatic Speaker Recognition (2009)
Tomi Kinnunen, Ville Hautamäki, Pasi Fränti
Abstract. State-of-the-art automatic speaker recognition systems use mel-frequency cepstral coefficients (MFCC) features to describe the spectral properties of speakers. In forensic phonetics, the...
On Factors Affecting MFCC-Based Speaker Recognition Accuracy (2009)
Juhani Saastamoinen, Zdenek Fiedler, Tomi Kinnunen, Pasi Fränti
We evaluate the accuracy of an MFCC-based speaker recognition method. We analyse the recognition results using speech signal from everyday life environments. We study the mismatch effects of...
Tomi Kinnunen, Juhani Saastamoinen, Ville Hautamäki, Mikko Vinni, Pasi Fränti
Gaussian mixture model- universal background model (GMM-UBM) is a standard reference classifier in speaker verification. We have recently proposed a simplified model using vector quantization...
Field Evaluation of Text-Dependent Speaker Recognition in an Access Control Application (2009)
Harsh Gupta, Ville Hautamäki, Tomi Kinnunen, Pasi Fränti
Vector quantization (VQ) is a widely used matching algorithm for text-independent speaker recognition. In this paper we study the use of text-dependent speaker recognition in practical access control...
Automatic Speaker Recognition for Series 60 Mobile Devices (2008)
Juhani Saastamoinen, Evgeny Karpov, Ville Hautamäki, Pasi Fränti
Mobile phone environment and other devices that require low power consumption, restrict the computations of DSP processors to fixed point arithmetic only. This is a common practise based on cost and...
Data reduction of large vector graphics (2008)
Alexander Kolesnikov, Pasi Fränti
Fast algorithm for joint near-optimal approximation of multiple polygonal curves is proposed. It is based on iterative reduced search dynamic programming introduced earlier for the min-ε problem of...
Polygonal approximation of closed contours (2008)
Abstract. Optimal approximation of closed curves differs from the case of open curve in the sense that the location of the starting point must also be determined. Straightforward exhaustive search...
Clustering by Principal Curve with Tree Structure (2008)
Ioan Cleju, Pasi Fränti, Xiaolin Wu
Abstract—Data clustering is intensively used in signal processing in tasks such as multimedia compression, segmentation and pattern matching. In this work we extend the use of principal curves in...
Automatic Speaker Recognition for Series 60 Mobile Devices (2008)
Juhani Saastamoinen, Evgeny Karpov, Ville Hautamäki, Pasi Fränti
Mobile phone environment and other devices that require low power consumption, restrict the computations of DSP processors to fixed point arithmetic only. This is a common practise based on cost and...
Is Speech Data Clustered? - Statistical Analysis of Cepstral Features (2008)
Tomi Kinnunen, Ismo Kärkkäinen, Pasi Fränti
Speech analysis applications are typically based on short-term spectral analysis of the speech signal. Feature extraction process outputs one feature vector per frame. The features are further...
Maximum a posteriori adaptation of the centroid model for speaker verification (2008)
Ville Hautamäki, Tomi Kinnunen, Ismo Kärkkäinen, Juhani Saastamoinen, Marko Tuononen, Pasi Fränti
Abstract—Maximum a posteriori adapted Gaussian mixture model (GMM-MAP) is widely used in speaker verification. GMMs have three sets of parameters to be adapted: means, covariances, and weights....
Compression of line drawing images using Hough transform for exploiting global dependencies (2007)
Pasi Fränti, Eugene I. Ageenko, Heikki Kälviäinen, Saku Kukkonen
: A two-stage method is proposed for compressing bi-level line drawing images. In the first stage line elements are extracted from the image using the Hough transform. In the second stage the...
Aritmetic Compression of Weighted Finite Automata (2007)
Karel Culik and the first author have demonstrated how Weighted Finite Automata (WFA) provide a strong tool for image compression [1, 2, 3, 4]. In the present article we introduce an improved method...
Context model automata for text compression (2007)
Timo Hatakka, Pasi Fränti, Pasi Fränti
Finite-state automata offer an alternative approach to implement context models where the states in the automata cannot in general be assigned by a single context. Despite the potential of this...
Pasi Fränti, Eugene Ageenko, Saku Kukkonen, Heikki Kälviäinen
On the use of Hough transform for context-based image compression
Morphological Reconstruction of Semantic Layers in Map Images (2006)
Alexey Podlasov, Eugene Ageenko, Pasi Fränti
Map images are composed of semantic layers depicted in arbitrary color. Color separation is often needed to divide the image into layers for storage and processing. Separation can result in severe...
Clustering based on principal curve (2005)
Ioan Cleju, Pasi Fränti, Xiaolin Wu
Abstract. Clustering algorithms are intensively used in the image analysis field in compression, segmentation, recognition and other tasks. In this work we present a new approach in clustering vector...
Accuracy of MFCC-Based Speaker Recognition in Series 60 Device (2005)
Juhani Saastamoinen, Evgeny Karpov, Ville Hautamäki, Pasi Fränti
A fixed point implementation of speaker recognition based on MFCC signal processing is considered. We analyze the numerical error of the MFCC and its effect on the recognition accuracy. Techniques to...
Kuopio, Finland Speaker Clustering in Speech Recognition (2005)
Olga Grebenskaya, Tomi Kinnunen, Pasi Fränti
The paper presents a combination of speaker and speech recognition techniques aiming to improve speech recognition rates. This combination is done by clustering the speaker models created from the...
Compression of Map Images for Real-Time Applications (2004)
Pasi Fränti, Pasi Fra Nti, Eugene Ageenko, Sami Gröhn, Pavel Kopylov, Sami Gro Hn, ...
Digital maps can be stored and distributed electronically using compressed raster image formats. We introduce a storage system for the map images that supports compact storage size, decompression of...
Fast Pairwise Nearest Neighbor Based Algorithm for Multilevel Thresholding (2003)
We propose a fast pairwise nearest neighbor (PNN)- based O(N log N) time algorithm for multilevel nonparametric thresholding, where N denotes the size of the image histogram. The proposed PNN-based...
On the Fusion of Dissimilarity-Based Classifiers for Speaker Identification (2003)
Tomi Kinnunen, Ville Hautamäki, Pasi Fränti
In this work, we describe a speaker identification system that uses multiple supplementary information sources for computing a combined match score for the unknown speaker. Each speaker profile in...
A speaker pruning algorithm for real-time speaker identification (2003)
Tomi Kinnunen, Evgeny Karpov, Pasi Fränti
Abstract. Speaker identification is a computationally expensive task. In this work, we propose an iterative speaker pruning algorithm for speeding up the identification in the context of real-time...
Self-Adaptive Genetic Algorithm for Clustering (2003)
Juha Kivijärvi, Pasi Fränti, Olli Nevalainen
Clustering is a hard combinatorial problem which has many applications in science and practice. Genetic algorithms (GAs) have turned out to be very effective in solving the clustering problem....
Pasi Fränti, Eugene Ageenko, Saku Kukkonen
In a hybrid raster/vector system, two representations of the image are stored. Digitized raster image preserves the original drawing in its exact visual form, whereas additional vector data can be...
Context-Based Compression Of Binary Images In Parallel (2002)
Eugene Ageenko, Martti Forsell, Pasi Fränti
this paper, we attack the above problem and present a method for the context-based compression of binary images in parallel. Our approach is to divide the compression problem into several smaller...
Pasi Fränti, Pasi Fra Nti, Eugene Ageenko, Saku Kukkonen
In a hybrid raster/vector system, two representations of the image are stored. Digitized raster image preserves the original drawing in its exact visual form, whereas additional vector data can be...
Map Image Compression for Real-Time Applications (2002)
Pasi Fränti, Eugene Ageenko, Pavel Kopylov, Sami Gröhn, Florian Berger
Digital maps can be stored and distributed electronically using compressed raster image formats. We introduce a storage system for the map images that supports compact storage size, decompression of...
Evaluation of Compression Methods for Digital Map Images (2002)
Pasi Fränti, Pavel Kopylov, Eugene Ageenko
Digital maps can be stored and distributed electronically using compressed raster image formats. In this paper, we evaluate the most suitable compression methods for storing digital map images. We...
Practical Methods for Speeding-Up the Pairwise Nearest Neighbor Method (2001)
Olli Virmajoki, Pasi Fränti, Timo Kaukoranta
The pairwise nearest neighbor (PNN) method is a simple and well-known method for codebook generation in vector quantization. In its exact form, it provides a good-quality codebook but at the cost of...
A Fast Exact GLA Based on Code Vector Activity Detection (2000)
Timo Kaukoranta, Pasi Fränti, Olli Nevalainen
This paper introduces a new method for reducing the number of distance calculations in the generalized Lloyd algorithm (GLA), which is a widely used method to construct a codebook in vector...
Lossless Compression Of Large Binary Images In Digital Spatial Libraries (2000)
A method for lossless compression of large binary images is proposed for applications where spatial access to the image is needed. The method utilizes the advantages of (1) variable-size context...
Content-based matching of line-drawing images using the Hough transform (2000)
Pasi Fränti, Alexey Mednonogov, Ville Kyrki, Heikki Kälviäinen
We intro duc two novel methods for cr tentbasedmatc hing of line-drawing images. The methods are based on the Hough transform (HT),whic h is used to extrac global line features in an image. The...
Binary Vector Quantizer Design Using Soft Centroids (1999)
Pasi Fränti, Pasi Frak Nti, Timo Kaukoranta
Soft centroids method is proposed for binary vector quantizer design. Instead of using binary centroids, the codevectors can take any real value between one and zero during the codebook generation...
Vector Quantization by Lazy Pairwise Nearest Neighbor Method (1999)
Timo Kaukoranta, Pasi Fränti, Olli Nevalainen
The Pairwise Nearest Neighbor (PNN) algorithm is a well-known method for the codebook construction in vector quantization, and for the clustering of data sets. The algorithm has simple structure and...
Lossless and near-lossless compression of line-drawing images using Hough transform (1998)
Pasi Fränti, Eugene I. Ageenko, Saku Kukkonen
: Two novel methods for compressing bi-level line-drawing images are proposed. The main idea is the use of the Hough transform (HT). In the first phase, HT is applied for extracting line segments...
Eugene I. Ageenko, Eugene I. Ageenko, Pasi Fränti, Pasi Frnti
: Main objectives of an engineering document management (EDM) system are outlined. Existing image compression algorithms (eg. G4 and JBIG) offer efficient solutions to the storage problem but do not...
On the splitting method for VQ codebook generation (1997)
Pasi Fränti, Timo Kaukoranta, Olli Nevalainen
in[ 1] , integrated with a local repartitioning phase at each step of the algorithm. 1.
Genetic Algorithms for Large Scale Clustering Problems (1997)
Pasi Fränti, Juha Kivijärvi, Timo Kaukoranta, Olli Nevalainen
We consider the clustering problem in a case where the distances of elements are metric and both the number of attributes and the number of the clusters are large.
Block Truncation Coding with Entropy Coding (1993)
: Block Truncation Coding (BTC) is a simple and fast image compression algorithm wich achieves constant bit rate of 2.0 bits per pixel. The method is however suboptimal. In the present paper we...
A Fast and Efficient Compression Method for Binary Images (1993)
The use of arithmetic coding for binary image compression achieves a high compression ratio while the running time remains rather slow. A composite modelling method presented in this paper reduces...
A Two-Stage Modelling Method for Compressing Binary Images by Arithmetic Coding (1992)
: A two-stage modelling schema to be used together with arithmetic coding is proposed. Main motivation of the work has been the relatively slow operation of arithmetic coding. The new modelling...
Fast and memory efficient implementation of the exact PNN (1990)
Pasi Fränti, Timo Kaukoranta, Day-fann Shen, Kuo-shu Chang
Abstract—Straightforward implementation of the exact pairwise nearest neighbor (PNN) algorithm takes @ Q A time, where is the number of training vectors. This is rather slow in practical...
Speaker Discriminative Weighting Method for VQ-based Speaker identification
We consider the matching function in vector quantization based speaker identification system. The model of a speaker is a codebook generated from the set of feature vectors from the speakers voice...