Table 2:

Baseline results of ASR trained only on children’s speech (91 hours).

Model	WER
GMM-HMM Monophone	54.53%
GMM-HMM Triphone	36.96%
GMM-HMM Triphone LDA+MLLT	32.79%
GMM-HMM Triphone LDA+MLLT+SAT	24.55%
GMM-HMM Triphone LDA+MLLT+SAT + VTLN	25.66%
Hybrid DNN-HMM	35.97%
Hybrid DNN-HMM + VTLN	32.72%
Hybrid DNN-HMM + LDA+MLLT+SAT	21.31%
Hybrid DNN-HMM + LDA+MLLT+SAT + VTLN	21.82%
Hybrid DNN-HMM + online i-vector (speaker)	28.03%
Hybrid DNN-HMM + online i-vector (utterance)	26.59%
Hybrid DNN-HMM + offline i-vector (utterance)	25.53%