Audiovisual articulatory inversion based on Gaussian Mixture Model (GMM) Akustik ve görsel veriler kullanilarak artikülatörlerin hareketlerinin Gauss Karişim Modellere (GKM) dayali kestirimi

ÖZBEK İ. Y., Demirekler M.

18th IEEE Signal Processing and Communications Applications Conference, SIU 2010, Diyarbakır, Türkiye, 22 - 24 Nisan 2010, ss.380-382, (Tam Metin Bildiri) identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/siu.2010.5653987
  • Basıldığı Şehir: Diyarbakır
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.380-382
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

In this study, we examined articulatory inversion using audiovisual information based on Gaussian Mixture Model (GMM). In this method the joint distribution of the articulatory movement and audio (and/or visual) data are modelled via a mixture of Gaussians. The conditional expected value of the GMM is used as regression function between the audio (and/orvisual) and articulatory spaces. We also examined various fusion methods in order to combine acoustic and visual information in articulatory inversion. The fusion methods improve the performance of articulatory inversion. ©2010 IEEE.