Patents Examined by Matthew H Baker
  • Patent number: 10303768
    Abstract: Technologies to detect persuasive multimedia content by using affective and semantic concepts extracted from the audio-visual content as well as the sentiment of associated comments are disclosed. The multimedia content is analyzed and compared with a persuasiveness model.
    Type: Grant
    Filed: October 2, 2015
    Date of Patent: May 28, 2019
    Assignee: SRI International
    Inventors: Ajay Divakaran, Behjat Siddiquie, David Chisholm, Elizabeth Shriberg
  • Patent number: 10304457
    Abstract: According to one embodiment, a transcription support system supports transcription work to convert voice data to text. The system includes a first storage unit configured to store therein the voice data; a playback unit configured to play back the voice data; a second storage unit configured to store therein voice indices, each of which associates a character string obtained from a voice recognition process with voice positional information, for which the voice positional information is indicative of a temporal position in the voice data and corresponds to the character string; a text creating unit that creates the text in response to an operation input of a user; and an estimation unit configured to estimate already-transcribed voice positional information indicative of a position at which the creation of the text is completed in the voice data based on the voice indices.
    Type: Grant
    Filed: March 15, 2012
    Date of Patent: May 28, 2019
    Assignee: KABUSHIKI KAISHA TOSHIBA
    Inventors: Hirokazu Suzuki, Nobuhiro Shimogori, Tomoo Ikeda, Kouji Ueno, Osamu Nishiyama, Manabu Nagao
  • Patent number: 10275712
    Abstract: A method providing an answer to at least one analytical question containing at least one table or at least one chart is provided. The method may include receiving an input question. The method may also include extracting a plurality of information from the input question based on a natural language analysis. The method may further include forming a well-defined sentence. The method may include extracting at least one table or at least one chart associated with the input question. The method may include forming at least one mathematical equation. The method may also include solving the at least one mathematical equation. The method may include determining the answer to the input question in natural language based on the solved at least one mathematical equation. The method may further include narrating the determined answer to the input question in natural language.
    Type: Grant
    Filed: June 21, 2016
    Date of Patent: April 30, 2019
    Assignee: International Business Machines Corporation
    Inventors: Sandesh Bhat, Joy Mustafi
  • Patent number: 10275713
    Abstract: A method providing an answer to at least one analytical question containing at least one table or at least one chart is provided. The method may include receiving an input question. The method may also include extracting a plurality of information from the input question based on a natural language analysis. The method may further include forming a well-defined sentence. The method may include extracting at least one table or at least one chart associated with the input question. The method may include forming at least one mathematical equation. The method may also include solving the at least one mathematical equation. The method may include determining the answer to the input question in natural language based on the solved at least one mathematical equation. The method may further include narrating the determined answer to the input question in natural language.
    Type: Grant
    Filed: June 21, 2016
    Date of Patent: April 30, 2019
    Assignee: International Business Machines Corporation
    Inventors: Sandesh Bhat, Joy Mustafi
  • Patent number: 10224056
    Abstract: A speech-based system includes a local device in a user premises and a network-based control service that directs the local device to perform actions for a user. The control service may specify a first action that is to be performed upon detection by the local device of a stimulus. In some cases, performing the first action may rely on the availability of network communications with the control service or with another service. In these cases, the control service also specifies a second, fallback action that does not rely upon network communications. Upon detecting the stimulus, the local device performs the first action if network communications are available. If network communications are not available, the local device performs the second, fallback action.
    Type: Grant
    Filed: December 17, 2013
    Date of Patent: March 5, 2019
    Assignee: Amazon Technologies, Inc.
    Inventors: Fred Torok, Rohan Mutagi, Vikram Kumar Gundeti, Frederic Johan Georges Deramat
  • Patent number: 10204619
    Abstract: Methods, systems, and apparatus are described that receive audio data for an utterance. Association data is accessed that indicates associations between data corresponding to uncorrupted audio segments, and data corresponding to corrupted versions of the uncorrupted audio segments, where the associations are determined before receiving the audio data for the utterance. Using the association data and the received audio data for the utterance, data corresponding to at least one uncorrupted audio segment is selected. A transcription of the utterance is determined based on the selected data corresponding to the at least one uncorrupted audio segment.
    Type: Grant
    Filed: February 22, 2016
    Date of Patent: February 12, 2019
    Assignee: Google LLC
    Inventors: Olivier Siohan, Pedro J. Moreno Mengibar
  • Patent number: 10199037
    Abstract: A reduced latency system for automatic speech recognition (ASR). The system can use certain feature values describing the state of ASR processing to estimate how far a lowest scoring node for an audio frame is from a potential node likely be part of the Viterbi path. The system can then adjust its beam width in a manner likely to encompass the node likely to be on the Viterbi path, thus pruning unnecessary nodes and reducing latency. The feature values and estimated distances may be based on a set of training data, where the system identifies specific nodes on the Viterbi path and determines what feature values correspond to what desired beam widths. Trained models or other data may be created at training and used at runtime to dynamically adjust the beam width, as well as other settings such as threshold number of active nodes.
    Type: Grant
    Filed: June 29, 2016
    Date of Patent: February 5, 2019
    Assignee: Amazon Technologies, Inc.
    Inventors: Denis Sergeyevich Filimonov, Yuan Shangguan
  • Patent number: 10176809
    Abstract: Systems and methods for compressing and decompressing audio data are described. A server may receive input audio data corresponding to a spoken utterance from a speech-controlled device. The server performs speech processing on the input audio data to determine a spoken command and spoken solicited content. The server may then communication with a device associated with an application to obtain output audio data. The server may compress the output audio data by removing portions of audio data therefrom at regular intervals (e.g., milliseconds). The server may then send the compressed output audio data and instructions for decompressing the compressed output audio data to the speech-controlled device. Further, a speech-capturing device may also compress audio data corresponding to an utterance. The speech-capturing device may compress based on various factors, including the speech characteristics of the speaking user.
    Type: Grant
    Filed: September 29, 2016
    Date of Patent: January 8, 2019
    Assignee: Amazon Technologies, Inc.
    Inventor: Adrien Hubert Marie Piérard
  • Patent number: 10109272
    Abstract: According to one embodiment, an apparatus for training a neural network acoustic model includes a calculating unit, a clustering unit, and a sharing unit. The calculating unit calculates, based on training data including a training speech and a labeled phoneme state, scores of phoneme states different from the labeled phoneme state. The clustering unit clusters a phoneme state whose score is larger than a predetermined threshold and the labeled phoneme state. The sharing unit shares probability of the labeled phoneme state by the clustered phoneme states. The training unit trains the neural network acoustic model based on the training speech and the clustered phoneme states.
    Type: Grant
    Filed: September 12, 2016
    Date of Patent: October 23, 2018
    Assignee: Kabushiki Kaisha Toshiba
    Inventors: Huifeng Zhu, Yan Deng, Pei Ding, Kun Yong, Jie Hao
  • Patent number: 10089976
    Abstract: A regional monitoring system can include a plurality of voice sensing units each of which incorporates speech recognition circuitry. In response to recognizing a verbal command at a unit, a coded representation, or token, along with location information, can be transmitted to a system control apparatus. Upon receipt of the token, the control apparatus can carry out the requested command or provide requested information.
    Type: Grant
    Filed: October 14, 2013
    Date of Patent: October 2, 2018
    Assignee: HONEYWELL INTERNATIONAL INC.
    Inventors: Eric Oh, Kenneth L. Addy, David S. Zakrewski
  • Patent number: 10043527
    Abstract: A method for generating a psychoacoustic model from an audio signal transforms a block of samples of an audio signal into a frequency spectrum comprising frequency components. From this frequency spectrum, it derives group masking energies. These group masking energies each correspond to a group of neighboring frequency components in the frequency spectrum. For a group of frequency components, the method allocates the group masking energy to the frequency components in the group in proportion to energy of the frequency components within the group to provide adapted mask energies for the frequency components within the group, the adapted mask energies providing masking thresholds for the psychoacoustic model of the audio signal.
    Type: Grant
    Filed: July 18, 2016
    Date of Patent: August 7, 2018
    Assignee: Digimarc Corporation
    Inventors: Aparna R. Gurijala, Shankar Thagadur Shivappa, Ravi K. Sharma, Brett A. Bradley
  • Patent number: 10026412
    Abstract: The present application discloses a pulse encoding method for use by a voice encoder. An initial number of coding bits is determined according to number of pulses in an input signal. An index value is calculated according to information of the pulses in the input signal. The index value is compared with an adjustment threshold value. If the index value is smaller than the adjustment threshold value, the index value is encoded using a reduced number of encoding bits. If the index value is not smaller than the adjustment threshold value, the index value plus an offset value is encoded using the initial number of encoding bits. The reduced number of encoding bits is smaller than the initial number of encoding bits, and the offset value is greater than or equal to the adjustment threshold value.
    Type: Grant
    Filed: May 11, 2016
    Date of Patent: July 17, 2018
    Assignee: HUAWEI TECHNOLOGIES CO., LTD.
    Inventors: Fuwei Ma, Dejun Zhang, Minjie Xie, Qing Zhang
  • Patent number: 7962330
    Abstract: An apparatus for automatic dissection of segmented audio signals, wherein at least one information signal for identifying programs included in said audio signals and for identifying contents included in said programs. Content detection device detects programs and contents belonging to the respective programs in the information signal. Program weighting device weights each program includes in the information signal based on the contents of the respective program detected by the content detection device. Program ranking device indentifies programmers of the same category and ranking said programs based on a weighting result for each program provided by the program weighting device.
    Type: Grant
    Filed: November 10, 2004
    Date of Patent: June 14, 2011
    Assignee: Sony Deutschland GmbH
    Inventors: Silke Goronzy, Thomas Kemp, Ralf Kompe, Yin Hay Lam, Krzysztof Marasek, Raquel Tato
  • Patent number: 7957953
    Abstract: A weighted linear word alignment model linearly combines weighted features to score a word alignment for a bilingual, aligned pair of text fragments. The features are each weighted by a feature weight. One of the features is a word association metric, which may be generated from surface statistics.
    Type: Grant
    Filed: October 3, 2005
    Date of Patent: June 7, 2011
    Assignee: Microsoft Corporation
    Inventor: Robert C. Moore
  • Patent number: 7930168
    Abstract: An advanced model that includes new processes is provided for use as a component of an effective disfluency identifier. The disfluency identifier tags edited words in transcribed speech. A speech recognition unit in combination with a part-of-speech tagger, a disfluency identifier, and a parser form a natural language system that helps machines properly interpret spoken utterances.
    Type: Grant
    Filed: October 4, 2005
    Date of Patent: April 19, 2011
    Assignee: Robert Bosch GmbH
    Inventors: Fuliang Weng, Qi Zhang
  • Patent number: 7853445
    Abstract: A system for identifying deception within a text includes a processor for receiving and processing a text file. The processor includes a deception indicator tag analyzer for inserting into the text file at least one deception indicator tag that identifies a potentially deceptive word or phrase within the text file, and an interpreter for interpreting the at least one deception indicator tag to determine a distribution of potentially deceptive word or phrases within the text file and generating deception likelihood data based upon the density or distribution of potentially deceptive word or phrases within the text file.
    Type: Grant
    Filed: December 8, 2005
    Date of Patent: December 14, 2010
    Assignee: Deception Discovery Technologies LLC
    Inventors: Joan C. Bachenko, Michael J. Schonwetter
  • Patent number: 7809548
    Abstract: The present invention provides a method of processing at least one natural language text using a graph. The method includes determining a plurality of text units based upon the natural language text, associating the plurality of text units with a plurality of graph nodes, and determining at least one connecting relation between at least two of the plurality of text units. The method also includes associating the at least one connecting relation with at least one graph edge connecting at least two of the plurality of graph nodes and determining a plurality of rankings associated with the plurality of graph nodes based upon the at least one graph edge. The method can also include a graphical visualization of at least one important text unit in a natural language text or collection of texts. Methods for word sense disambiguation, keyword extraction, and sentence extraction are also provided.
    Type: Grant
    Filed: March 9, 2005
    Date of Patent: October 5, 2010
    Assignee: University of North Texas
    Inventors: Rada Mihalcea, Paul Tarau
  • Patent number: 7809579
    Abstract: Polyphonic signals are used to create a main signal, typically a mono signal, and a side signal. A number of encoding schemes for the side signal are provided. Each encoding scheme is characterized by a set of sub-frames of different lengths. The total length of the sub-frames corresponds to the length of the encoding frame of the encoding scheme. The encoding scheme to be used on the side signal is selected dependent on the present signal content of the polyphonic signals. In a preferred embodiment, a side residual signal is created as the difference between the side signal and the main signal scaled with a balance factor. The balance factor is selected to minimize the side residual signal. The optimized side residual signal and the balance factor are encoded and provided as encoding parameters representing the side signal.
    Type: Grant
    Filed: December 15, 2004
    Date of Patent: October 5, 2010
    Assignee: Telefonaktiebolaget LM Ericsson (publ)
    Inventors: Stefan Bruhn, Ingemar Johansson, Anisse Taleb, Daniel Enström
  • Patent number: 7788106
    Abstract: The present invention is based on the finding that an efficient code for encoding information values can be derived, when two or more information values are grouped in a tuple in a tuple order and when an encoding rule is used, that assigns the same code word to tuples having identical information values in different orders and that does derive an order information, indicating the tuple order, and when the code word is output in association with the order information.
    Type: Grant
    Filed: October 14, 2005
    Date of Patent: August 31, 2010
    Assignee: Fraunhofer-Gesellschaft zur Foerderung der Angewandten Forschung E.V.
    Inventors: Ralph Sperschneider, Jürgen Herre, Karsten Linzmeier, Johannes Hilpert
  • Patent number: 7788085
    Abstract: String replacement is performed in text using linguistic processing. The linguistic processing identifies the existence of direct or indirect links between the string to be replaced and other strings in the text. Morphological, syntactic, anaphoric, or semantic inconsistencies, which are introduced in strings with the identified direct or indirect links to the string that is to be replaced are detected and corrected.
    Type: Grant
    Filed: December 17, 2004
    Date of Patent: August 31, 2010
    Assignee: Xerox Corporation
    Inventors: Caroline Brun, Herve Dejean, Caroline Hagege