首页|Patent Application Titled "Systems And Methods For Correcting Automatic Speech R ecognition Errors" Published Online (USPTO 20240153490)
Patent Application Titled "Systems And Methods For Correcting Automatic Speech R ecognition Errors" Published Online (USPTO 20240153490)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-According to news reporting originatin g from Washington, D.C., by NewsRx journalists, a patent application by the inve ntors Bonilla, Guadalupe (Hyattsville, MD, US); Can, Aysu Ezen (Cary, NC, US); C ritzer, Meredith Leigh (Chesterfield, VA, US); Edwards, Joshua (Philadelphia, PA , US); Khanagha, Vahid (Rockville, MD, US); Lin, Alexander (Arlington, VA, US); Maiman, Tyler (Melville, NY, US); Mossoba, Michael (Great Falls, VA, US); Qiu, F eng (Newark, DE, US); Rodriguez, Mia (Arlington, VA, US), filed on January 17, 2 024, was made available online on May 9, 2024. No assignee for this patent application has been made. Reporters obtained the following quote from the background information supplied by the inventors: "Automatic speech recognition is a widely used technique for c onverting speech to text. Many organizations rely on ASR for customer service an alysis and training of machine learning models. For example, organizations utili ze ASR for transcribing customers' telephone conversations with customer service agents for use in training agents how to provide more efficient services to cus tomers. Additionally, organizations rely on ASR for identification of fraudulent activity, such as a fraudster attempting to speak with a cus-tomer service agent as if he or she was a legitimate customer. Thus, the accuracy of ASR model tran scription is important for purposes of both training and any downstream modeling.
NETL
NSTL
