Adaptation of male acoustic models on the basis of language-independent female voice data

A method for generating proxy acoustic models for use in automatic speech recognition, comprising the steps of: (A), acoustic models from speech material, which has been obtained via the microphone of male speakers of a first language training, and (B) in step (a) trained acoustic models in response to said independent voice data from female speakers of a second language to be adapted to generate proxy acoustic models for use during the duration of the speech recognition of an utterance of a female speaker of the first language.

[1]  Biing-Hwang Juang,et al.  Fundamentals of speech recognition , 1993, Prentice Hall signal processing series.