Radiotekhnika
Publishing house Radiotekhnika

"Publishing house Radiotekhnika":
scientific and technical literature.
Books and journals of publishing houses: IPRZHR, RS-PRESS, SCIENCE-PRESS


Тел.: +7 (495) 625-9241

 

Recognition of pronunciation of decimal digits on the background of noise using autocorrelation portraits

Keywords:

A.I. Armer – Ph. D. (Eng.), Associate Professor, Department of Applied Mathematics and Informatics, Ulyanovsk State Technical University. E-mail: a.armer@mail.ru M.V. Batrakov – Student, Ulyanovsk State Technical University. E-mail: m_a_x_73@mail.ru V.R. Krasheninnikov – Dr. Sc. (Eng.), Professor, Head of Department of Applied Mathematics and Informatics, Ulyanovsk State Technical University. E-mail: kvrulstu@mail.ru N.A. Krasheninnikova – Ph. D. (Eng.), Associate Professor, Head of Department of English Language для профессиональной деятельности, Ulyanovsk State University. E-mail: kna.73@mail.ru


Currently speech technology is widely used in various applications. For example, in aviation and navy messages can be transmitted to humans in their natural language. The presence of intense noise complicates the recognition of voice messages, so a very limited vocabulary can be used. This enables the automated recognition of the transmitted information using various algorithms of word recognition on the background of noise. In this paper we consider the case when the information is transmitted as a sequence of decimal numbers (0, 1, …, 9). Thus, the problem lies in the recognition of digitized pronunciations of these numbers on the background of strong acoustic and other noise. It is proposed to conduct such recognition by converting voice signals into specific images, namely autocorrelation portraits. The recognition is conducted by comparing the portrait of an identifiable word with its model portrait. To improve the probability of correct recognition the model library optimization is used.
References:

 

  1. Krasheninnikov V.R., Armer A.I., Krasheninnikova N.A., KHvostov A.V.Raspoznavanie rechevykh komand na fone intensivnykh shumov s pomoshhju avtokorreljacionnykh portretov // Naukoemkie tekhnologii. 2007. T. 8. № 9. S. 65−76.
  2. Krasheninnikov V.R., Armer A.I., Krasheninnikova N.A., Kuznecov V.V., KHvostov A.V. Nekotorye zadachi, svjazannye s raspoznavaniem rechevykh komand na fone intensivnykh shumov // Infokommunikacionnye tekhnologii. Samara. 2008. № 1. S. 72−75.
  3. Krasheninnikov V.R., Armer A.I., Kuznecov V.V. Autocorrelated Images and Search for Distance between them in Speech Commands Recognition // Pattern Recognition and Image Analysis. 2008. V. 18. № 4. P. 663−666.
  4. Vasilev K.K., Krasheninnikov V.R. Statisticheskijj analiz izobrazhenijj. Uljanovsk: UlGTU. 2014. 214 s.
  5. Krasheninnikov V.R., Armer A.I., KHvostov A.V. KHarakteristiki izmenchivosti dlitelnostejj rechevykh signalov // Trudy XIIMezhdunar. nauchno-tekhnich. konf. «Radiolokacija navigacija svjaz» RLNC–2006. Voronezh. 2006. S. 927−933.
  6. Krasheninnikov V.R., Armer A.I.Sinicyn I.N. Model izmenchivosti rechevykh komand // Naukoemkie tekhnologii. 2007. № 9. S. 56−64.
  7. Krasheninnikov V.R.Krasheninnikova N.A.Kuznecov V.V., Lebedeva E.JU. Optimization of dictionary and model library for recognition of speech commands // Pattern Recognition and Image Analysis. 2011. V. 21. № 3. P. 505−507.

 

© Издательство «РАДИОТЕХНИКА», 2004-2017            Тел.: (495) 625-9241                   Designed by [SWAP]Studio