Optical character recognition training data generation for neural networks by parsing page description language jobs

A computing device, comprising:

• one or more processors;
• and
• non-transitory data storage storing at least computer-readable instructions that, when executed by the one or more processors, cause the computing device to perform tasks comprising:
• receiving training documents comprising text, wherein the training documents are specified using a page description language (PDL);
• parsing the PDL for the training documents to determine a plurality of training data items, each training data item of the plurality of training data items comprising a training label related to text within the training documents and location information comprising a bounding box indicating a location of text within the training documents that is related to the training label;
• training an artificial neural network (ANN) to recognize text in documents using the plurality of training data items and training input, the training input comprising the training documents;
• after training the ANN, receiving a request to predict text in application documents that differ from the training documents, the application documents comprising second text;
• determining, by applying the trained ANN to the application documents, a prediction of the second text;
• and
• after determining the prediction of the second text, providing information related to the second text as output.

An article of manufacture comprising non-transitory data storage storing at least computer-readable instructions that, when executed by one or more processors of a computing device, cause the computing device to perform tasks comprising:

• receiving training documents comprising text, wherein the training documents are specified using a page description language (PDL);
• parsing the PDL of the training documents to determine a plurality of training data items, each training data item of the plurality of training data items comprising a training label related to text within the training documents and location information comprising a bounding box indicating a location of text within the training documents that is related to the training label;
• training an artificial neural network (ANN) to recognize text in documents using the plurality of training data items and training input, the training input comprising the training documents;
• after training the ANN, receiving a request to predict text in application documents that differ from the training documents, the application documents comprising second text;
• determining, by applying the trained ANN to the application documents, a prediction of the second text;
• and
• after determining the prediction of the second text, providing information related to the second text as output.