Support Vector Machines (SVMs) have become a popular classification tool. Because of their theoretical robustness they offer improvements in pattern classification applications. This paper describes an approach of producing a N-best list of hypotheses for the needs of phoneme recognition, using a Least Squares Support Vector Machine classifier (LS-SVM) and generate the corresponding N-best list...

The phase of a signal conveys critical information for signal interpretation. However, signal phase extraction is not a straightforward process, mainly due to the discontinuities appearing in the phase spectrum. Previously, it was shown that for the application of audio classification, in case features of the processed phase spectrum are combined with magnitude related features they provide hig...

In the presence of noise and sensor mismatch condition performance of a conventional automatic Hindi speech recognizer starts to degrade, while we human being are able to segregate, focus and recognize the target speech. In this paper, we have used auditory based feature extraction procedure Gammatone frequency cepstral coefficient (GFCC) for Hindi phoneme classification. To distinguish vowels ...

In this thesis, Support Vector Machines (SVMs) are applied to the problem of phoneme classification. Given a sequence of acoustic observations and 40 phoneme targets, the task is to classify each observation to one of these targets. Since this task involves multiple classes, one of the main hurdles SVMs must overcome is to extend the inherently binary SVMs to the multi-class case. To do this, s...

In this study four statistical grapheme-to-phoneme (G2P) conversion methods for canonical German are compared. The G2P models differ in terms of usage of morphologic information and of phoneme history (left context) information. In order to evaluate our models we introduce two measures, namely mean normalized Levenshtein distance for classification accuracy and conditional relative entropy for ...

The robustness of classification of isolated phoneme segments using generative classifiers is investigated for the acoustic waveform and PLP speech representations. Probabilistic PCA is used to fit a density to each phoneme class followed by maximum likelihood classification. The results show that although PLP performs best in quiet conditions, as the SNR decreases below 0dB acoustic waveforms ...

Our former study [1] has shown that maximum likelihood (ML) based frame selection, which selects reliable frames from a high resolution along the time axis, helps to improve the discrimination between phonemes. In this paper, we present our recent research on single frame selection for a phoneme classification task. A new single selection, which only selects one frame for one state in an Hidden...

In this paper, we carry out two experiments on the TIMIT speech corpus with bidirectional and unidirectional Long Short Term Memory (LSTM) networks. In the first experiment (framewise phoneme classification) we find that bidirectional LSTM outperforms both unidirectional LSTM and conventional Recurrent Neural Networks (RNNs). In the second (phoneme recognition) we find that a hybrid BLSTM-HMM s...

To investigate the interaction in speech perception of auditory information and lexical knowledge (in particular, knowledge of which phonetic sequences are words), acoustic continua varying in voice onset time were constructed so that for each acoustic continuum, one of the two possible phonetic categorizations made a word and the other did not. For example, one continuum ranged between the wor...

In Italian, effects of age of acquisition (AoA) have been found in object naming, semantic categorization of words and lexical decision, but not in word naming (reading aloud). The lack of an AoA effect in Italian word naming is replicated in Experiment 1 which involved reading aloud two-syllable words which all have regular spelling-sound correspondences and regular stress patterns. Studies of...