Artificial intelligence

Scientific journal

ISSN 2710-1673

ONLINE: ISSN 2710-1681

Search by:

Year of publication
Author name
Paper title

Adaptation to New Announcer Voice for Spontaneous Speech from AKUEM Speech Corpus

Yukhymenko O.1, Pilipenko V.2, Seliukh R.1
1 International Research and Training Centre for Information Technologies and Systems
2 Speech Science and Technology Department, International Research and Training Center of Information Technologies and Systems «CyberMova»

Full text (PDF)

UDC: 004.934
Publication Language: Ukrainian
Stuc. intelekt. 2013; 18(3):284-292

Abstract: This article is devoted to the problems of adaptation to new announcer voice for speech recognition systems. The results of three adaptation experiments based on AKUEM speech corpus are described. Comparison with previous experiments are discussed.

Keywords: models of phonemes, speaker adaptation, recognition, linear transformations, classes of regression, training

References:

  1. Sazhok M. Adaptatsija akustychnykh modelej fonem do holosu dyktora dlya pofonemnogo rozpiznavannya izolyovanykh sliv ukrajinskoji movy / M. Sazhok, R. Selyukh , O. Yukhymenko // Shtuchnyj intelekt. – Donetsk, 2009. – No 4. – s. 230-233.
  2. Sazhok M. Adaptatsija do holosu dyktora na osnovi gendernozalezhnykh akustychnykh modelej fonem dlya ukrajinskoji movy / M. Sazhok, R. Selyukh , O. Yukhymenko. – Obroblennya sygnaliv i zobrazhen ta rozpiznavannya obraziv : Desyata Vseukrajinska mizhnarodna konferenciya. – Kyiv, 2010. – s. 59-62.
  3. Young S.J. HTK Book, version 3.1 / Young S.J. [et al]. – Cambridge University, 2002. – 355 p.
  4. Stvorennya akustychnoho korpusu ukrajinskoho movlennya / [N.B. Vasyl’eva, V.V. Pylypenko, O.M. Radutskyj et al]. – Desyata Vseukrajinska mizhnarodna konferenciya. – Kyiv, 2010. – s. 55-58.

View full text (PDF)