Company Filing History:
Years Active: 2018-2019
Title: Przemyslaw Tomaszewski: Innovator in Key Phrase Detection Technologies
Introduction
Przemyslaw Tomaszewski is a notable inventor based in Reda, Poland. He has made significant contributions to the field of voice recognition technology, particularly in key phrase detection techniques. With a total of 2 patents, his work has implications for low power applications in voice-activated systems.
Latest Patents
Tomaszewski's latest patents focus on linear scoring for low power wake on voice. The key phrase detection techniques he discusses involve performing a vectorized operation on a multiple element acoustic score vector for a current time instance. This includes a single state rejection model score and scores for a multiple state key phrase model. Additionally, it incorporates a multiple element state score vector for a previous time instance, which includes a previous state score for the single state rejection model and previous state scores for the multiple state key phrase model. The process generates a multiple element score summation vector, followed by a second vectorized operation to determine a multiple element state score vector for the current time instance. Ultimately, this vector may be evaluated to ascertain whether the received audio input includes a key phrase corresponding to the multiple state key phrase model.
Career Highlights
Przemyslaw Tomaszewski is currently employed at Intel Corporation, where he continues to innovate in the field of voice recognition technology. His work is instrumental in advancing the capabilities of low power voice-activated systems.
Collaborations
Tomaszewski collaborates with talented individuals such as Tobias Bocklet and Tomasz Dorau, contributing to a dynamic and innovative work environment.
Conclusion
Przemyslaw Tomaszewski's contributions to key phrase detection technologies highlight his role as a significant inventor in the field. His patents and work at Intel Corporation demonstrate his commitment to advancing voice recognition technology.