INTERSPEECH 2015
16th Annual Conference of the International Speech Communication Association

Dresden, Germany
September 6-10, 2015

Exploiting i-Vector Posterior Covariances for Short-Duration Language Recognition

Sandro Cumani (1), Oldřich Plchot (2), Radek Fér (2)

(1) Politecnico di Torino, Italy
(2) Brno University of Technology, Czech Republic

Linear models in i-vector space have shown to be an effective solution not only for speaker identification, but also for language recognition. The i-vector extraction process, however, is affected by several factors, such as noise level, the acoustic content of the utterance and the duration of the spoken segments. These factors influence both the i-vector estimate and its uncertainty, represented by the i-vector posterior covariance matrix. Modeling of i-vector uncertainty with Probabilistic Linear Discriminant Analysis has shown to be effective for short-duration speaker identification. This paper extends the approach to language recognition, analyzing the effects of i-vector covariances on a state-of-the-art Gaussian classifier, and proposes an effective solution for the reduction of the average detection cost (Cavg) for short segments.

Full Paper

Bibliographic reference.  Cumani, Sandro / Plchot, Oldřich / Fér, Radek (2015): "Exploiting i-vector posterior covariances for short-duration language recognition", In INTERSPEECH-2015, 1002-1006.