101 results about "Speech recognition performance" patented technology

Techniques are disclosed for generating and using sparse representation features to improve speech recognition performance. In particular, principles of the invention provide sparse representation exemplar-based recognition techniques. For example, a method comprises the following steps. A test vector and a training data set associated with a speech recognition system are obtained. A subset of the training data set is selected. The test vector is mapped with the selected subset of the training data set as a linear combination that is weighted by a sparseness constraint such that a new test feature set is formed wherein the training data set is moved more closely to the test vector subject to the sparseness constraint. An acoustic model is trained on the new test feature set. The acoustic model trained on the new test feature set may be used to decode user speech input to the speech recognition system.

A distributed speech recognition system and method thereof in accordance with the present invention enables a word and a natural language to be recognized using detection of a pause period in a speech period in an inputted speech signal, and various groups of recognition vocabulary (for example, a home speech recognition vocabulary, a telematics vocabulary for a vehicle, a vocabulary for call center, and so forth) to be processed in the same speech recognition system by determining the recognition vocabulary required by a corresponding terminal using an identifier of the terminal since various terminals require various speech recognition targets. In addition, various types of channel distortion occurring due to the type of terminal and the recognition environment are minimized by adapting them to a speech database model using a channel estimation method so that the speech recognition performance is enhanced.

The invention provides a voice recognizing method and a voice recognizing device, as well as a neural network training method and a neural network training device, and relates to the technical field of artificial intelligence. The neural network training method comprises the following steps: acquiring sample data, wherein the sample data comprises mixed voice frequency spectra and labelling phoneme of the mixed voice frequency spectra; extracting the target voice frequency spectrum from the mixed voice frequency spectra through a first sub-network; carrying out adaptive conversion on the target voice frequency spectrum through a second sub-network, so that middle transition representation is obtained; carrying out phoneme recognition based on the middle transition representation through athird sub-network; and according to the phoneme recognition result and the labelling phoneme, updating the parameters of the first sub-network, the second sub-network and the third sub-network. With the technical scheme, the voice recognition performance under the complicated interference sound condition can be promoted.

Techniques are disclosed for generating and using sparse representation features to improve speech recognition performance. In particular, principles of the invention provide sparse representation exemplar-based recognition techniques. For example, a method comprises the following steps. A test vector and a training data set associated with a speech recognition system are obtained. A subset of the training data set is selected. The test vector is mapped with the selected subset of the training data set as a linear combination that is weighted by a sparseness constraint such that a new test feature set is formed wherein the training data set is moved more closely to the test vector subject to the sparseness constraint. An acoustic model is trained on the new test feature set. The acoustic model trained on the new test feature set may be used to decode user speech input to the speech recognition system.