Method and apparatus for feature domain joint channel and additive noise compensation
AssigneeMatsushita Electric Industrial Co., Ltd.
InventorsYounes Souilmi, Luca Rigazio, Patrick Nguyen, Jean-Claude Junqua
Life of this patent
Estimating the term…
Abstract
A method for performing noise adaptation of a target speech signal input to a speech recognition system, where the target speech signal contains both additive and convolutional noises. The method includes estimating an additive noise bias and a convolutional noise bias; in the target speech signal; and jointly compensating the target speech signal for the additive and convolutional noise biases in a feature domain.
Claims26
A method for performing noise adaptation of a target speech signal input to a speech recognition system, said target speech signal containing both additive and convolutional noises, said method comprising:
- estimating an additive noise bias and a convolutional noise bias;
- in said target speech signal;
- and
- jointly compensating the target speech signal for the additive and convolutional noise biases in a feature domain.
A speech recognition apparatus including noise adaptation of a target speech signal input to said apparatus, said target speech signal containing both additive and convolutional noises, said apparatus configured to:
- estimate an additive noise bias and a convolutional noise bias in said target speech signal;
- and
- jointly compensate the target speech signal for the additive and convolutional noise biases in a feature domain.
17 An apparatus in accordance with claim 16 further configured to determine noise compensation matrix JN as:
- 28 J N = F † 1 H tar · S + N tar ,
- where F† is a decorrelating transform and Htar·S+Ntar is the noisy target speech signal.
26 An apparatus in accordance with claim 25 further configured to said model in the spectral domain from a model in the cepstral domain inverse cepstral operator C−1(·), wherein C−1(x)=exp(F−1x), F is a matrix, F† is a decorrelating transform and the inverse conjugate of F, and FF†=I, where I is an identity matrix.
Dependencies are inferred from the claim text.