Integrating text inputs for training and adapting neural network transducer ASR models
AssigneeInternational Business Machines Corporation
InventorsSamuel Thomas, Hong-Kwang Kuo, Brian E. D. Kingsbury, George Andrei Saon, Gakuto Kurata
Life of this patent
Estimating the term…
Abstract
A processor-implemented method trains an automatic speech recognition system using speech data and text data. A computer device receives speech data, and generates a spectrogram based on the speech data. The computing device receives text data associated with an entire corpus of text data, and generates a textogram based upon the text data. The computing device trains an automatic speech recognition system using the spectrogram and the textogram.
Claims20
A method of using a computing device to train an automatic speech recognition system using speech data and text data, the method comprising:
- receiving, by a computing device, speech data;
- generating, by the computing device, a spectrogram based on the speech data;
- receiving, by the computing device, text data associated with an entire corpus of text data;
- generating, by the computing device, a textogram based upon the text data;
- and
- training, by the computing device, an automatic speech recognition system using the spectrogram and the textogram.
A computer program product for training an automatic speech recognition system using speech data and text data, wherein the computer program product comprises a non-transitory computer readable storage device having program instructions embodied therewith, the program instructions readable and executable by a computer to perform a method comprising:
- receiving speech data;
- generating a spectrogram based on the speech data;
- receiving text data associated with an entire corpus of text data;
- generating a textogram based upon the text data;
- and
- training an automatic speech recognition system using the spectrogram and the textogram.
A computer system comprising one or more processors, one or more computer readable memories, and one or more computer readable non-transitory storage mediums, and program instructions stored on at least one of the one or more computer readable non-transitory storage mediums for execution by at least one of the one or more processors via at least one of the one or more computer readable memories, the stored program instructions executed to perform a method comprising:
- receiving speech data;
- generating a spectrogram based on the speech data;
- receiving text data associated with an entire corpus of text data;
- generating a textogram based upon the text data;
- and
- training an automatic speech recognition system using the spectrogram and the textogram.
Dependencies are inferred from the claim text.