| model,16-3-H90-1060,bq | In addition , combination of the <term> training speakers </term> is done by averaging the <term> statistics > </term> of <term> independently trained models </term> rather than the usual pooling of all the <term> speech data </term> from many <term> speakers </term> prior to <term> training </term> . | #17050 In addition, combination of the training speakers is done by averaging the statistics> of independently trained models rather than the usual pooling of all the speech data from many speakers prior to training. |
