B
Bjorn Hoffmeister
Researcher at Amazon.com
Publications - 87
Citations - 2468
Bjorn Hoffmeister is an academic researcher from Amazon.com. The author has contributed to research in topics: Acoustic model & Word error rate. The author has an hindex of 25, co-authored 85 publications receiving 2151 citations. Previous affiliations of Bjorn Hoffmeister include Apple Inc. & Wilmington University.
Papers
More filters
Proceedings ArticleDOI
Multi-task learning and Weighted Cross-entropy for DNN-based Keyword Spotting
Sankaran Panchapagesan,Ming Sun,Aparna Khare,Spyros Matsoukas,Arindam Mandal,Bjorn Hoffmeister,Shiv Naga Prasad Vitaladevuni +6 more
TL;DR: It is shown that the combination of 3 techniques LVCSR-initialization, multi-task training and weighted cross-entropy gives the best results, with significantly lower False Alarm Rate than the LV CSR- initialization technique alone, across a wide range of Miss Rates.
Patent
Accepting voice commands based on user identity
TL;DR: In this article, the authors present techniques for determining when to perform an action associated with a voice command and when to disregard the voice command, and reference an identity of a user that utters a command when making this determination.
Journal ArticleDOI
Speech Processing for Digital Home Assistants: Combining signal processing with deep-learning techniques
Reinhold Haeb-Umbach,Shinji Watanabe,Tomohiro Nakatani,Michiel Bacchiani,Bjorn Hoffmeister,Michael L. Seltzer,Heiga Zen,Mehrez Souden +7 more
TL;DR: The purpose of this article is to describe, in a way that is amenable to the nonspecialist, the key speech processing algorithms that enable reliable, fully hands-free speech interaction with digital home assistants.
Proceedings Article
The RWTH aachen university open source speech recognition system.
David Rybach,Christian Gollan,Georg Heigold,Bjorn Hoffmeister,Jonas Lööf,Ralf Schlüter,Hermann Ney +6 more
TL;DR: The toolkit includes state of the art speech recognition technology for acoustic model training and decoding, and a finite state automata library, and an efficient tree search decoder are notable components.
Patent
Speech model retrieval in distributed speech recognition systems
TL;DR: In this paper, features for managing the use of speech recognition models and data in automated speech recognition systems are disclosed, including pre-caching and pre-processing models and statistics.