Search results for "Speech recognition"
showing 10 items of 357 documents
Cardiovascular and respiratory variability during orthostatic and mental stress: A comparison of entropy estimators
2017
The aim of this study is to characterize cardiovascular and respiratory signals during orthostatic and mental stress as reflected in indices of entropy and complexity, providing a comparison between the performance of different estimators. To this end, the heart rate variability, systolic blood pressure, diastolic blood pressure and respiration time series were extracted from the recordings of 61 healthy volunteers undergoing a protocol consisting of supine rest, head-up tilt test and mental arithmetic task. The analysis was performed in the information domain using measures of entropy and conditional entropy, estimated through model-based (linear) and model-free (binning, nearest neighbor)…
Perception and replication of planar sonic gestures
2012
As tables, boards, and walls become surfaces where interaction can be supported by auditory displays, it becomes important to know how accurately and effectively a spatial gesture can be rendered by means of an array of loudspeakers embedded in the surface. Two experiments were designed and performed to assess: (i) how sequences of sound pulses are perceived as gestures when the pulses are distributed in space and time along a line; (ii) how the timing of pulses affects the perceived and reproduced continuity of sequences; and (iii) how effectively a second parallel row of speakers can extend sonic gestures to a two-dimensional space. Results show that azimuthal trajectories can be effectiv…
Ventricular Fibrillation and Tachycardia detection from surface ECG using time-frequency representation images as input dataset for machine learning
2017
Parameter-less ventricular fibrillation detection with time-frequency representation.Time-frequency representations are treated as images for a classifier.A comparison for four classifiers demonstrates the validity of the proposed method.The proposed technique could be applied to any signal and research field.This is a novel approach to signal analysis. Background and objectiveTo safely select the proper therapy for Ventricullar Fibrillation (VF) is essential to distinct it correctly from Ventricular Tachycardia (VT) and other rhythms. Provided that the required therapy would not be the same, an erroneous detection might lead to serious injuries to the patient or even cause Ventricular Fibr…
The computation of word associations
2002
It is shown that basic language processes such as the production of free word associations and the generation of synonyms can be simulated using statistical models that analyze the distribution of words in large text corpora. According to the law of association by contiguity, the acquisition of word associations can be explained by Hebbian learning. The free word associations as produced by subjects on presentation of single stimulus words can thus be predicted by applying first-order statistics to the frequencies of word co-occurrences as observed in texts. The generation of synonyms can also be conducted on co-occurrence data but requires second-order statistics. The reason is that synony…
A method for the time-varying nonlinear prediction of complex nonstationary biomedical signals
2009
A method to perform time-varying (TV) nonlinear prediction of biomedical signals in the presence of nonstationarity is presented in this paper. The method is based on identification of TV autoregressive models through expansion of the TV coefficients onto a set of basis functions and on k -nearest neighbor local linear approximation to perform nonlinear prediction. The approach provides reasonable nonlinear prediction even for TV deterministic chaotic signals, which has been a daunting task to date. Moreover, the method is used in conjunction with a TV surrogate method to provide statistical validation that the presence of nonlinearity is not due to nonstationarity itself. The approach is t…
Subtle Increases in Interletter Spacing Facilitate the Encoding of Words during Normal Reading
2012
BackgroundSeveral recent studies have revealed that words presented with a small increase in interletter spacing are identified faster than words presented with the default interletter spacing (i.e., w a t e r faster than water). Modeling work has shown that this advantage occurs at an early encoding level. Given the implications of this finding for the ease of reading in the new digital era, here we examined whether the beneficial effect of small increases in interletter spacing can be generalized to a normal reading situation.MethodologyWe conducted an experiment in which the participant's eyes were monitored when reading sentences varying in interletter spacing: i) sentences were present…
Study time effects in recognition memory.
2004
We empirically tested the assumption that study time increases recognition accuracy because the storage of information is better when study time is longer as Shiffrin and colleagues have reported, an assumption common to parallel models of recognition. In the present study with 123 subjects, we examined the effect of item strength on four measures: hit rate, false alarm rate, d′, and β, for a single-word recognition task with longer study times than those usually used in the literature. Analysis indicated significant increase for hit rate and d′ and a decrease in false alarm rate, as one goes from weak to stronger study conditions, and a change in ln(β) when study time is greater than 1 se…
Learning the structure of HMM's through grammatical inference techniques
2002
A technique is described in which all the components of a hidden Markov model are learnt from training speech data. The structure or topology of the model (i.e. the number of states and the actual transitions) is obtained by means of an error-correcting grammatical inference algorithm (ECGI). This structure is then reduced by using an appropriate state pruning criterion. The statistical parameters that are associated with the obtained topology are estimated from the same training data by means of the standard Baum-Welch algorithm. Experimental results showing the applicability of this technique to speech recognition are presented. >
Effort in Semi-Automatized Subtitling Processes
2020
The presented study investigates the impact of automatic speech recognition (ASR) and assisting scripts on effort during transcription and translation processes, two main subprocesses of interlingual subtitling. Applying keylogging and eye tracking, this study takes a first look at how the integration of ASR impacts these subprocesses. 12 professional subtitlers and 13 translation students were recorded performing two intralingual transcriptions and three translation tasks to evaluate the impact on temporal, technical, and cognitive effort, and split-attention. Measures include editing time, visit count and duration, insertions, and deletions. The main findings show that, in both tasks, ASR…
A Semiotic Approach to the Design of Non-speech Sounds
2006
In the field of auditory display there is currently a lack of theoretical support for the design of non-speech sounds as elements of a user interface. Sound design methods are often based on ad hoc choices or the personal preferences of the designer. A method is proposed in this paper based on a semiotic approach to the design of non-speech sounds. In this approach, the design process is conceptualised by referring to structural semiotics, acknowledging the unique qualities of non-speech sounds, as a mode of conveying information. This method is based on a rich use scenario presented to a design panel. A case study where the design method has been applied is presented and evaluated. Finally…