N2 - The same speech sounds (phones) produced by different speakers can sometimes exhibit significant differences. Therefore, it is essential to use algorithms compensating these differences in ASR systems. Speaker clustering is an attractive solution to the compensation problem, as it does not require long utterances or high computational effort at the recognition stage. The report proposes a clustering method based solely on adaptation of UBM model weights. This solution has turned out to be effective even when using a very short utterance. The obtained improvement of frame recognition quality measured by means of frame error rate is over 5%. It is noteworthy that this improvement concerns all vowels, even though the clustering discussed in this report was based only on the phoneme a. This indicates a strong correlation between the articulation of different vowels, which is probably related to the size of the vocal tract. L1 - http://www.czasopisma.pan.pl/Content/101332/PDF/aoa-2016-0011.pdf L2 - http://www.czasopisma.pan.pl/Content/101332 PY - 2016 IS - No 1 EP - 118 DO - 10.1515/aoa-2016-0011 KW - automatic speech recognition KW - interindividual difference compensation KW - speaker clustering KW - universal background model KW - GMM weighting factor adaptation A1 - Hossa, Robert A1 - Makowski, Ryszard PB - Polish Academy of Sciences, Institute of Fundamental Technological Research, Committee on Acoustics VL - vol. 41 DA - 2016 T1 - An Effective Speaker Clustering Method using UBMand Ultra-Short Training Utterances SP - 107 UR - http://www.czasopisma.pan.pl/dlibra/publication/edition/101332 T2 - Archives of Acoustics