EURASIP Journal on Audio, Speech, and Music Processing

Table 1 Statistics of the datasets used for training: number of words in the dictionary and amount of speakers and data

From: Exploiting foreign resources for DNN-based ASR

Dataset	Dict.	No. of speakers	Data (h)
MP-FR (French)	11 k	106	19
MP-GE (German)	16 k	75	18
ESTER (French)	203 k	669	57
BREF (French)	13 k	218	114

Back to article page