1,721,009 research outputs found
On the use of channel-attentive MFCC for robust recognition of partially corrupted speech
Full text linkThis letter proposes a channel-attentive Mel frequency cepstral coefficient (CAMFCC) method to improve the utilization of uncorrupted or more reliable frequency bands for robust speech recognition. This method obtains a channel attention matrix by reliability estimation of Mel filter bank channels, and both the input Mel frequency cepstral coefficients and the mean vectors of hidden Markov models are corrected using the channel attention matrix at the output probability calculation of the Viterbi decoding. Experimental results on the TIDIGITS database corrupted by various band-selective noises indicated that the proposed CAMFCC method utilizes the uncorrupted partial frequency bands better than a multiband method, resolving the limitation of noise localization caused by the fixed boundaries of the multiband approach
Segmental reliability weighting for robust recognition of partly corrupted speech
No full textA new method for the recognition of speech that is partly corrupted by noise is proposed. For this purpose, the reliability of each speech segment is measured based on a log likelihood ratio. A segment that has higher reliability is given more importance in the decision making by a modified Viterbi algorithm. Experimental results showed that the method significantly improves the performance of isolated word recognition under various burst noise situations
Missing data techniques using voicing probability dor robust automatic speech recognition
No full textThe authors propose a new method for detecting missing data by utilising voicing probability under a missing data theory. With the same level of distortion, people fail to recognise vowels more frequently than consonants. From this observation, we Propose that consonants should not be classified into missing data. The experimental results showed that our method significantly improves the performance for isolated word recognition under various noise environments
Speech Extraction from Harmonic Interference using GMM based Classification of Separated Spectra
No full textNEGATIVELY CHARGED STATE OF ATOMIC-HYDROGEN IN N-TYPE GAAS
No full textIt is demonstrated that atomic hydrogen drifts as a negatively charged state in n-type GaAs and the high electric field strongly affects the dissociation of the hydrogen-donor complex. During reverse-bias-anneal experiments on the Schottky diode, it is confirmed that a negatively charged hydrogen atom is accelerated out of the high-field region and that there is a dissociation-frequency region independent of the anneal temperature. In the dissociation-frequency region dependent on the anneal temperature, the first-order kinetics gives rise to the dissociation energy for the release of the hydrogen-Si donor complex. The dissociation energies are dependent on the applied bias voltage and are in the range of 1.79 to 1.2 eV. Atomic hydrogen in plasma-hydrogenated Si-doped n-type GaAs is proposed to be negatively charged with the gain of free electrons and passivates the Si donor, and also the hydrogen or the electron of the hydrogen-Si donor complex to be easily released by the electric field.
Ethylene Polymerization with Ni-Diimine/MAO : Number of Active Centers and Propagation Rate Cinstant
No full text
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
