Voice based manual image review

A system for manual-based image review, comprising:

• a display screen for displaying an image associated with a plurality of keyword utterances, wherein the plurality of keyword utterances includes different keyword utterances;
• a user interface for displaying for a user, a prompt for the user to utter at least one keyword utterance among the plurality of keyword utterances in a first area of the image displayed in the display screen and another prompt for the user to utter at least one other keyword utterance among the plurality of keyword utterances in another area of the image displayed in the display screen;
• and
• an audio device for capturing and processing by natural language processing (NLP), audio of the at least one keyword utterance and the at least one other keyword utterance when uttered by the user through an audio device operable to operate with the NLP, for display of the at least one keyword utterance and the at least one other keyword utterance as text in the respective first area and the another area of the image, in response to processing by NLP of the audio, wherein the NLP is operable to understand the at least one keyword utterance and post NLP results resulting from the capturing and the processing by the NLP;
• wherein the NLP understands the at least one keyword utterance and the at least one other keyword utterance uttered by the user posts NLP results to a back office after the processing of the audio;
• and
• the NLP comprises at least one of:
• NLP based on a hybrid sequence-to-sequence approach, and NLP processing review and override based on confidence analysis.

A computer program product for facilitating manual-based image review of images, the computer program product comprising one or more non-transitory computer readable storage media and program instructions collectively stored on the one or more non-transitory computer readable storage media, the program instructions comprising program instructions to:

• associate an image with a plurality of keyword utterances, the image displayable in a display screen of a computing device, the plurality of keyword utterances including different keyword utterances;
• display for a user, a prompt for the user to utter at least one keyword utterance among the plurality of keyword utterances in a first area of the image displayed in the display screen and another prompt for the user to utter at least one other keyword utterance among the plurality of keyword utterances in another area of the image displayed in the display screen;
• and
• capture and process by natural language processing (NLP), audio of the at least one keyword utterance and the at least one other keyword utterance when uttered by the user through an audio device operable to operate with the NLP, for display of the at least one keyword utterance and the at least one other keyword utterance as text in the respective first area and the another area of the image, in response to processing by NLP of the audio, wherein the NLP is operable to understand the at least one keyword utterance and post NLP results resulting from the capturing and the processing by the NLP;
• wherein the NLP understands the at least one keyword utterance and the at least one other keyword utterance uttered by the user posts NLP results to a back office after the processing of the audio;
• and
• the NLP comprises at least one of:
• NLP based on a hybrid sequence-to-sequence approach, and NLP processing review and override based on confidence analysis.