Connectionist Bench (Vowel Recognition - Deterding Data) Data Set
Below are papers that cite this data set, with context shown.
Papers were automatically harvested and associated with this data set, in collaboration with Rexa.info.
Return to Connectionist Bench (Vowel Recognition - Deterding Data) data set page.
M. Layton and M. J. F Gales. CAMBRIDGE UNIVERSITY ENGINEERING DEPARTMENT Maximum Margin Training of Generative Kernels. CUED. 2004.
the number of iterations required for convergence increased significantly (34 iterations were required for C = 10, whereas 1,084 were required for C = 0.1). 15 5.2 Deterding data The Deterding dataset  consists of the steady-state portion of 11 vowels in British English, spoken in the context of h*d. The recorded speech samples were low-pass filtered at 4.7KHz before being digitised at 10KHz
Matthew Brand. Pattern discovery via entropy minimization. MERL -- A MITSUBISHI ELECTRIC RESEARCH LABORATORY. 1998.
DA circumvents this local optimum and finds a model which generalizes even better. 9.2 Radial basis function networks (RBFNs) Vowel classification: We obtained the British English vowel recognition dataset from the CMU Neural-Bench Archive. Each example of a vowel is characterized by LPC coefficients from two time-frames. This dataset has been treated several times using perceptrons, neural networks,