Speech Recognition
AssigneeAlibaba Group Holding Limited
InventorShaofei Xue
Life of this patent
Estimating the term…
Abstract
A speech recognition method includes clustering feature vectors of training data to obtain clustered feature vectors of training data performing interpolation calculation on feature vectors of data to be recognized using the clustered feature vectors of training data, and inputting the feature vectors of data to be recognized after the interpolation calculation into a speech recognition model to adaptively adjust the speech recognition model. The techniques of the present disclosure improve speech recognition accuracy and adaptive processing efficiency.
Claims20
A method comprising:
- clustering, by one or more processors of a computing device, feature vectors of training data to obtain clustered feature vectors of training data;
- performing interpolation calculation on feature vectors of data to be recognized using the clustered feature vectors of training data;
- and
- inputting the feature vectors of data to be recognized into a speech recognition model to adaptively adjust the speech recognition model.
A device comprising:
- a clustering module configured to cluster feature vectors of training data to obtain clustered feature vectors of training data;
- an interpolation module configured to perform interpolation calculation on feature vectors of data to be recognized using the clustered feature vectors of training data;
- and
- an adaptive adjustment module configured to input the feature vectors of data to be recognized after the interpolation calculation into a speech recognition model to adaptively adjust the speech recognition model.
One or more memories stored thereon computer-readable instructions that, when executed by the one or more processors, cause the one or more processors to perform acts comprising:
- clustering, by one or more processors of a computing device, feature vectors of training data to obtain clustered feature vectors of training data;
- performing interpolation calculation on feature vectors of data to be recognized using the clustered feature vectors of training data;
- and
- inputting the feature vectors of data to be recognized into a speech recognition model to adaptively adjust the speech recognition model.
Dependencies are inferred from the claim text.