AUDIO-based Cloud-enabled Language and Intelligence Processing (AUDIO-CLIP)

Military operational tempo requires rapid development of highly accurate intelligence products from massive stores of data. Manual processing of data on this scale is infeasible. Automated processing systems have been developed to help analysts find the most relevant data to their current information requirements. This automated processing allows analysts to focus their efforts on performing higher level analysis. Unfortunately, most of these tools require the data to be in proper English. However, a significant percentage of the most fruitful data is in the native language of the areas of operation. Due to high error rates of automated transcription and translation tools, the majority of analytical tools are unable to produce meaningful results from this data. Therefore, DAC proposes to the AUDIO-based Cloud-enabled Language and Intelligence Processing (AUDIO-CLIP) system. The AUDIO-CLIP system will be focused on making all collected audio data available for real-time intelligence analysis and product generation. The algorithms within the system will focus on improving the automated generation of intelligence from foreign audio. AUDIO-CLIP will operate on foreign audio, transcriptions, and translations with the ultimate goal of improving the accuracy of the English transcripts for use by downstream analytical engines while also directly extracting intelligence at each step.