Trevor D Strohman

Title: Innovations by Trevor D. Strohman in Speech Recognition Technologies

Introduction

Trevor D. Strohman is an accomplished inventor located in Sunnyvale, California. With an impressive portfolio of 22 patents, he has made significant contributions to the fields of speech recognition and audio processing. His innovative work has been instrumental in advancing technologies at Google Inc., where he currently applies his expertise.

Latest Patents

Trevor's recent patents showcase his ability to align cutting-edge technology with practical applications. One of his notable patents, "Emitting Word Timings with End-to-End Models," describes a method for processing spoken utterances. This method employs placeholders to establish ground truth alignments for words within the audio data, facilitating enhanced accuracy in speech recognition by constraining attention in decoding processes.

Another significant patent, "Rare Word Recognition with LM-Aware MWER Training," introduces methods for generating refined hypotheses in speech recognition. This entails creating higher-order feature representations from acoustic frames and determining likelihood scores through a learnable fusion module. Such advancements aim to improve recognition accuracy for less common words by leveraging external language models.

Career Highlights

Throughout his career at Google Inc., Trevor has demonstrated a profound understanding of speech technologies. His dedication to research and innovation has not only resulted in numerous patents but also in the shaping of industry standards for speech recognition.

Collaborations

Trevor works alongside talented researchers, including notable coworkers Tara N. Sainath and Ruoming Pang. This collaborative environment fosters a dynamic exchange of ideas, contributing to the advancement of their projects and innovations in the field.

Conclusion

Trevor D. Strohman stands out as a pioneering inventor making significant strides in speech recognition technology. With a solid foundation of patents, collaborative efforts, and a commitment to innovation, he continues to impact the evolution of audio processing methodologies at Google Inc. His work not only enhances the capabilities of machines to understand human speech but also paves the way for future advancements in the field.