Voci Technologies Incorporated (Voci?) is partnering with Carnegie Mellon University (CMU) to develop and demonstrate a prototype Automated Spoken Language Recognition System (ASLRS). The ASLRS is specifically designed to enhance the usability of Speech to Speech (S2S) Machine Foreign Language Translation (MFLT) systems for the warfighter. The proposed ASLRS will leverage the Team?s existing language identification capabilities, experience, and expertise to fulfill the requirements of an efficient MFLT preprocessor. Best-in-class accuracy will be achieved using a combination of techniques and fusing the results. To meet the real-time requirements, a ground-breaking, patent-pending, multi-language phonetic dictionary capable of doing phonetic recognition in all 6 target languages in a single pass will be utilized. An open-set solution will be provided so that the ASLRS recognizes when an out-of-domain language is spoken. To ensure that the resulting ASLRS is generally applicable, it will be architected to be an open system, ensuring that it is inter-operable with existing MFLT solutions and that it supports the addition of new languages. To ensure that the system provides reliable results, even in noisy environment, the system will incorporate noise robust features. Finally, to address the shortcomings of existing solutions in real-world field conditions, the Team will integrate a learning capability into the ASLRS so that it can adapt to different accents and noise conditions that exist during field use. At the end of Phase II, the Team will demonstrate the prototype ASLRS on a mobile device (e.g., Android smartphone). We believe the final implementation will revolutionize S2S MFLT use in the field.
Keywords: (1) Automated Spoken Language Recognition, (2) Human Language Technology, (3) Language Identification (Lid), (4) Gender Identification, (5) Speaker Identification (Sid), (6) W
