V. V. Savchenko’s research while affiliated with Linguistic University of Nizhny Novgorod and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (6)


A Method of Measuring the Index of Acoustic Voice Quality Based on an Information-Theoretic Approach
  • Article

April 2018

·

24 Reads

·

5 Citations

Measurement Techniques

V. V. Savchenko

A new method is proposed for measurements of an index of acoustic voice quality using the Kulbaka–Leybler information metric. The primary benefit of this method consists in its high-level dynamic properties, as calculated by eliminating the problem of a small number of observational samples. A theoretical study of the method’s efficiency was conducted, and its conclusions were confirmed experimentally. It has been established that, in order to provide a sufficiently precise assessment of the voice quality of the speaker, a speech signal of duration 2–3 minutes is required.


Estimation of the Phonetic Speech Quality Using the Information Theoretic Approach

January 2018

·

8 Reads

·

8 Citations

Journal of Communications Technology and Electronics

The problem of estimation of the speech phonetic quality using a short fragment of the voice signal is formulated. The Kullback–Leibler minimum information discrimination principle is used to propose a novel criterion and develop an algorithm. The dynamic properties of the algorithm are analyzed. The algorithm is employed in the analysis of the functional state of speaker using the voice signal. It is demonstrated that the needed duration of the voice signal is several minutes only.


Study of the Stationarity of Random Time Series Using the Principle of the Information-Divergence Minimum

July 2017

·

6 Reads

·

7 Citations

Radiophysics and Quantum Electronics

Using the theoretic-information approach and the criterion of the information-divergence minimum in the Kullback–Leibler metric, we propose a new algorithm for checking the time series for stationarity in a broad sense. We consider an example of realizing this algorithm, study its dynamic characteristics, and give recommendations on its use under conditions of small samples.


Words phonetic decoding method with the suppression of background noise

July 2017

·

10 Reads

·

9 Citations

Journal of Communications Technology and Electronics

A new algorithm for the suppression of background noise in systems of recognition of voice commands is proposed based on phonetic decoding of words. The principle of the algorithm employs frequencydomain correction of quasi-stationary fragments of the signal of voice command using selective estimation of noise variance in pauses between syllables. The desired effect lies in a several-fold increase in the signal-to-noise ratio at the input of the decision unit. The theoretical results are proven with the aid of a physical experiment.


Enhancement of the noise immunity of a voice-activated robotics control system based on phonetic word decoding method

December 2016

·

19 Reads

·

10 Citations

Journal of Communications Technology and Electronics

The problem of protection of a voice-activated robotics control system against the acoustic noise similar to the speaker voice is investigated. A new algorithm for voice signal processing using the phonetic word decoding method and the (R+1) information element is proposed. The results of the experimental study of this algorithm are reported. It is shown that the developed algorithm reduces the false alarm ratio by and an order of magnitude or even more and retains high probability of correct voice recognition.


The Principle of the Information-Divergence Minimum in the Problem of Spectral Analysis of the Random Time Series Under the Condition of Small Observation Samples

October 2015

·

10 Reads

·

17 Citations

Radiophysics and Quantum Electronics

We consider the issue of small observation samples in the problem of spectral analysis of the random time series. It is proposed to solve the considered problem using the information-theoretic approach and a new algorithm based on the principle of minimum divergence of the cognominal spectral estimates yielded by the results of several independent observations in the Kullback–Leibler information metric. An example of a practical realization of the algorithm is considered and its asymptotic properties are studied.

Citations (5)


... Discussion of the results. Specialists are aware of the existence of a strong interdependence between the dynamics of the FF of a speech signal and its acoustic quality [16][17][18]. At observation intervals of several seconds, the FF of a high-quality signal, even if it varies, is within rather narrow limits (units of Hz). ...

Reference:

Method for Measuring the Indicator of Acoustic Quality of Audio Recordings Prepared for Registration and Processing in the Unified Biometric System
Estimation of the Phonetic Speech Quality Using the Information Theoretic Approach
  • Citing Article
  • January 2018

Journal of Communications Technology and Electronics

... In the ideal case where the PF is stabilized and does not increase as the speaker continues his speech, its QAS is characterized by the maximum value of the parameter Q(k) = 100%. Any false notes in the speech are refl ected in the program as a decrease in the parameter given by formula (7). The greater this decrease, the more pronounced the indicated false notes. ...

Words phonetic decoding method with the suppression of background noise
  • Citing Article
  • July 2017

Journal of Communications Technology and Electronics

... As a result of sequential frame processing, the sequence of sounds in the speech signal is determined. Based on the identified sequence of sounds, it is possible to artificially determine syllables and words in a speech signal [17]. To do this, an implicit Markov model of a sequence consisting of n-states is first constructed. ...

Study of the Stationarity of Random Time Series Using the Principle of the Information-Divergence Minimum
  • Citing Article
  • July 2017

Radiophysics and Quantum Electronics

... In the analyzed problem, W ≈ 12 dB. This gain is obtained, to the same extent, both for white noise and the noise similar to the noise of speech signal with "colored" power spectra [15]. The second type of noise has an intraperiodic fi ne (formant) structure that could not be distinguished from the useful signal and, hence, represents a currently unsolvable problem for the major part of available methods of measuring the PF [12]. ...

Enhancement of the noise immunity of a voice-activated robotics control system based on phonetic word decoding method
  • Citing Article
  • December 2016

Journal of Communications Technology and Electronics

... However, the indicated operating conditions are most typical of the ASA systems [8]. The problem becomes even more serious under the conditions of small samples of the data of observations for relatively short (fractions of a second) intervals of vocalization of speech signals [9]. This is why the problem of measurement of the PF in the real-time mode is an urgent scientifi c and engineering problem. ...

The Principle of the Information-Divergence Minimum in the Problem of Spectral Analysis of the Random Time Series Under the Condition of Small Observation Samples
  • Citing Article
  • October 2015

Radiophysics and Quantum Electronics