Speech to speech translation can benefit from translation of emphasis. We propose to use an intonation model to retrieve and transfer events associated with emphasis in the intonation. This model decomposes the F0 contour into basic intonation atoms using the matching pursuit algorithm. We investigate the role of these components in the perception of emphasis. Some of the most prominent local c...

The relationship between literature and politics is a reciprocal one. core theme of makes positive impact on individual society. Literature was considered as powerful tool to explore the moral, ethical social values which brought notable changes in socio – political sphere. In this context, Thirukkural most important universally accepted it reveals administrative system found erstwhile Tamil so...

Perceptual acoustic echo cancellers were developed by mainly considering human hearing thresholds of different acoustic frequencies in the past. In addition to the different frequency sensitivities, the human brain further analyzes sounds in terms of their spectral and temporal modulations. In this paper, we extend the perceptual normalized least mean square (P-NLMS) algorithm by adding a secon...

In this paper, an improved strategy for automated text dependent speaker identification system has been proposed in noisy environment. The identification process incorporates the Hidden Markov Model technique with cepstral based features. To remove the background noise from the source utterance, wiener filter has been used. Different speech pre-processing techniques such as start-end point dete...

Running Title: pragmatic use of language in pre-morbid adolescents Abstract At eleven years of age all children in a UK national birth cohort wrote short stories about the life they expected to be leading at age 25. Using a data linkage exercise, we identified those who later developed schizophrenia , affective psychosis , or other non-psychotic psychiatric disorders in later life based on the ...

Cepstral analysis is used to estimate the harmonics-to-noise ratio (HNR) in speech signals. The inverse Fourier transformed liftered cepstrum approximates a noise baseline from which the harmonics-to-noise ratio is estimated. The present study highlights the manner in which the cepstrum-based noise baseline estimate is obtained, essentially behaving like a moving average filter applied to the p...

Mel Frequency Cepstral Coefficients (MFCC) and Perceptual Linear Prediction (PLP) are the most popular acoustic features used in speech recognition. Often it depends on the task, which of the two methods leads to a better performance. In this work we develop acoustic features that combine the advantages of MFCC and PLP. Based on the observation that the techniques have many similarities, we rev...

We train a bank of complex filters that operates on the raw waveform and is fed into a convolutional neural network for end-to-end phone recognition. These time-domain filterbanks (TD-filterbanks) are initialized as an approximation of melfilterbanks, and then fine-tuned jointly with the remaining convolutional architecture. We perform phone recognition experiments on TIMIT and show that for se...