Patent
Block synchronous decoding
William H. Rockenbeck,JJ Odell +1 more
- 04 Mar 2003
194
TL;DR: In this article, a pattern recognition system and method for multi-state Hidden Markov Models (HMM) is described. But this method is not suitable for the use of HMM blocks.
read more
Abstract: A pattern recognition system and method are provided. Aspects of the invention are particularly useful in combination with multi-state Hidden Markov Models. Pattern recognition is effected by processing Hidden Markov Model Blocks. This block-processing allows the processor to perform more operations upon data while such data is in cache memory. By so increasing cache locality, aspects of the invention provide significantly improved pattern recognition speed.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Patent
Zero latency digital assistant
William F. Stasior,David A. Carson,Rohit Dasari,Yoon Kim +3 more
- 30 Aug 2016
TL;DR: In this article, a zero-latency digital assistant is implemented by capturing audio input from a microphone and using a first processor to write audio data representing the captured audio input to a circular memory buffer.
66
Patent
Parsimonious handling of word inflection via categorical stem + suffix N-gram language models
Jerome R. Bellegarda,Sibel Yaman +1 more
- 28 Aug 2015
TL;DR: In this paper, a system and processes for predicting words using a categorical stem and suffix word n-gram language model are disclosed for determining a stem probability using a stem language model and determining a suffix probability using suffix language model decoupled from the stem model.
63
Patent
Proactive assistance based on dialog communication between devices
Mathieu J. Martel,Thomas Deniau +1 more
- 31 May 2016
TL;DR: In this article, a text representation of speech contained in a portion of the stream of audio data can be generated and a process can determine whether the text representation contains information corresponding to one of a plurality of types of information.
61
Patent
Multilingual word prediction
Giulia Pagallo,Jerome R. Bellegarda,Vivek Kumar Rangarajan Sridhar,Randal J. Marsden,Justin S. Hogg +4 more
- 20 Sep 2016
TL;DR: In this article, a multilingual word prediction system was proposed, where a first candidate word of the plurality of candidate words corresponds to a first language of a plurality of languages and a second candidate word is a second language of languages different than the first language.
60
Patent
Data driven natural language event detection and classification
Jerome R. Bellegarda,Mathieu J. Martel,Alexandre Carlhian,Antoine Vinel,Jannes G. Dolfing,Sabrine Rekik,Markus Berget,Hafid Arras +7 more
- 25 Jul 2016
TL;DR: In this article, the authors present a system for operating a digital assistant that receives unstructured natural language information from at least one user and determines whether an agreement on an event is present in the UML information.
58
References
Single tree method for grammar directed, very large vocabulary speech recognizer
Richard Schwartz,Long Nguyen +1 more
TL;DR: The invention provides a method of large vocabulary speech recognition that employs a single tree-structured phonetic hidden Markov model (HMM) at each frame of a time-synchronous process, and phonetic context information is exploited, even before the complete context of a phoneme is known.
278
Patent
Speaker recognition device
TL;DR: A speaker recognition device for judging whether or not an unknown speaker is an authentic registered speaker himself/herself executes text verification using speaker independent speech recognition and speaker verification by comparison with a reference pattern of a password of a registered speaker as mentioned in this paper.
150
Unsupervised Analysis of Human Gestures
Tianshu Wang,Heung-Yeung Shum,Ying-Qing Xu,Nanning Zheng +3 more
- 24 Oct 2001
TL;DR: An approach for automatically segmenting sequences of natural activities into atomic sections and clustering them and extracting lexicon from these discrete sequences by using an algorithm named COMPRESSIVE is presented.
Time-first search for large vocabulary speech recognition
Tony Robinson,J. Christie +1 more
- 12 May 1998
TL;DR: A new search technique for large vocabulary speech recognition based on a stack decoder that proceeds time-first, that is partial path hypotheses are extended into the future in the inner loop and a tree walk over the lexicon is performed as an outer loop.
31
Patent
Speech recognition device
Takano Masaru
- 02 Dec 1997
TL;DR: In this article, a high performance speech recognition device with high recognition performance even for utterance with an additional word was provided by performing pattern-matching between a feature amount of an inputted speech and a standard pattern for every frame, selecting an optimum word sequence (optimum string), calculating its likelihood and deciding/outputting an optimum recognition result from the optimum string and the likelihood.
11
Related Papers (5)
Ken Kocienda
- 11 Aug 2008
Jong-Ho Shin,Jae-Do Kwak,Jong-Keun Youn +2 more
- 16 Jun 2008
Renaud J. Lecoeuche
- 10 Jan 2004