Method for Adapting a Codebook for Speech Recognition

A computer program product including a tangible computer-readable medium having computer code thereon for adapting a codebook for speech recognition, wherein the codebook is from a set of codebooks comprising a speaker-independent codebook and at least one speaker-dependent codebook, each codebook a set of Gaussian densities, the computer code comprising:

• (a) computer code for receiving a speech input;
• (b) computer code for determining a feature vector based on the received speech input;
• (c) computer code, for each of the Gaussian densities, for estimating a first mean vector using an expectation process and taking into account the determined feature vector;
• (d) computer code, for each of the Gaussian densities, for estimating a second mean vector using an Eigenvoice adaptation and taking into account the determined feature vector;
• and
• (e) computer code, for each of the Gaussian densities, for setting its mean vector to a convex combination of the first and the second mean vector.

An apparatus for adapting a codebook for speech recognition, wherein the codebook includes a set of Gaussian densities, the apparatus comprising:

• a receiver for receiving a speech input;
• a feature vector module for determining a feature vector based on the received speech input;
• a first estimation module for estimating, for each of the Gaussian densities, a first mean vector using an expectation maximization process and taking into account the determined feature vector;
• a second estimation module for estimating, for each of the Gaussian densities, a second mean vector using an Eigenvoice adaptation and taking into account the determined feature vector;
• and
• an adapter module, for each of the Gaussian densities, setting its mean vector to a convex combination of the first and the second mean vector.