Open AccessProceedings Article
Feature combination using linear discriminant analysis and its pitfalls.
Reads0
Chats0
TLDR
It is shown that the combination of acoustic features using LDA does not consistently lead to improvements in word error rate, and relative improvements inword error rate of up to 5% were observed for LDA-based combination of multiple acoustic features.Abstract:
In this paper, Linear Discriminant Analysis (LDA) is investigated with respect to the combination of different acoustic features for automatic speech recognition. It is shown that the combination of acoustic features using LDA does not consistently lead to improvements in word error rate. A detailed analysis of the recognition results on the Verbmobil (VM II) and on the English portion of the European Parliament Plenary Sessions (EPPS) corpus is given. This includes an independent analysis of the effect of the dimension of the input to LDA, the effect of strongly correlated input features, as well as a detailed numerical analysis of the generalized eigenvalue problem underlying LDA. Relative improvements in word error rate of up to 5% were observed for LDA-based combination of multiple acoustic features.read more
Citations
More filters
Proceedings ArticleDOI
Gammatone Features and Feature Combination for Large Vocabulary Speech Recognition
TL;DR: The gammatone features presented here lead to competitive results on the EPPS English task, and considerable improvements were obtained by subsequent combination to a number of standard acoustic features, i.e. MFCC, PLP, MF-PLP, and VTLN plus voicedness.
BookDOI
Handbook of Natural Language Processing and Machine Translation
TL;DR: This comprehensive handbook, written by leading experts in the field, details the groundbreaking research conducted under the breakthrough GALE program--The Global Autonomous Language Exploitation within the Defense Advanced Research Projects Agency (DARPA), while placing it in the context of previous research in the fields of natural language and signal processing, artificial intelligence and machine translation.
Dissertation
A log-linear discriminative modeling framework for speech recognition.
Georg Heigold,Hermann Ney +1 more
TL;DR: A log-linear modeling framework is established in the context of discriminative training criteria, with examples from continuous speech recognition, part-of-speech tagging, and handwriting recognition, and the focus will be on the theoretical and experimental comparison of different training algorithms.
Journal ArticleDOI
Combining Spectral Representations for Large-Vocabulary Continuous Speech Recognition
Giulia Garau,Steve Renals +1 more
TL;DR: The results indicate that combining conventional and pitch-synchronous acoustic feature sets using HLDA results in a consistent, significant decrease in word error rate across all three LVCSR tasks.
Proceedings ArticleDOI
Hierarchical Bottle Neck Features for LVCSR
TL;DR: Even though the hierarchical and bottle neck processing performs equally well, the combination of both topologies improves the system by 5% relative, and the MFCC baseline system is improved by up to 20% relative.
References
More filters
Journal ArticleDOI
Numerical Recipes in C: The Art of Scientific Computing
Mary C. Seiler,Fritz A. Seiler +1 more
Journal ArticleDOI
Perceptual linear predictive (PLP) analysis of speech
TL;DR: A new technique for the analysis of speech, the perceptual linear predictive (PLP) technique, which uses three concepts from the psychophysics of hearing to derive an estimate of the auditory spectrum, and yields a low-dimensional representation of speech.
Proceedings ArticleDOI
Linear discriminant analysis for improved large vocabulary continuous speech recognition
Reinhold Haeb-Umbach,Hermann Ney +1 more
TL;DR: The interaction of linear discriminant analysis (LDA) and a modeling approach using continuous Laplacian mixture density HMM is studied experimentally and the largest improvements in speech recognition could be obtained when the classes for the LDA transform were defined to be sub-phone units.
Journal ArticleDOI
The generalized Schur decomposition of an arbitrary pencil A–lB—robust software with error bounds and applications. Part I: theory and algorithms
James Demmel,Bo Kågström +1 more
TL;DR: Robust software with error bounds for computing the generalized Schur decomposition of an arbitrary matrix pencil A – λB (regular or singular) is presented.
Related Papers (5)
Heteroscedastic discriminant analysis and reduced rank HMMs for improved speech recognition
N. Kumar,Andreas G. Andreou +1 more