Patents by Inventor Michael Alan Picheny

Michael Alan Picheny has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 11183194
    Abstract: Aspects of the present disclosure describe techniques for identifying and recovering out-of-vocabulary words in transcripts of a voice data recording using word recognition models and word sub-unit recognition models. An example method generally includes receiving a voice data recording for transcription into a textual representation of the voice data recording. The voice data recording is transcribed into the textual representation using a word recognition model. An unknown word is identified in the textual representation, and the unknown word is reconstructed based on recognition of sub-units of the unknown word generated by a sub-unit recognition model. The textual representation of the voice data recording is modified by replacing the unknown word with the reconstruction of the unknown word, and the modified textual representation is output.
    Type: Grant
    Filed: September 13, 2019
    Date of Patent: November 23, 2021
    Assignee: International Business Machines Corporation
    Inventors: Samuel Thomas, Kartik Audhkhasi, Zoltan Tueske, Yinghui Huang, Michael Alan Picheny
  • Patent number: 11158303
    Abstract: In an approach to soft-forgetting training, one or more computer processors train a first model utilizing one or more training batches wherein each training batch of the one or more training batches comprises one or more blocks of information. The one or more computer processors, responsive to a completion of the training of the first model, initiate a training of a second model utilizing the one or more training batches. The one or more computer processors jitter a random block size for each block of information for each of the one or more training batches for the second model. The one or more computer processors unroll the second model over one or more non-overlapping contiguous jittered blocks of information. The one or more computer processors, responsive to the unrolling of the second model, reduce overfitting for the second model by applying twin regularization.
    Type: Grant
    Filed: August 27, 2019
    Date of Patent: October 26, 2021
    Assignee: International Business Machines Corporation
    Inventors: Kartik Audhkhasi, George Andrei Saon, Zoltan Tueske, Brian E. D. Kingsbury, Michael Alan Picheny
  • Publication number: 20210312906
    Abstract: An illustrative embodiment includes a method for training an end-to-end (E2E) spoken language understanding (SLU) system. The method includes receiving a training corpus comprising a set of text classified using one or more sets of semantic labels but unpaired with speech and using the set of unpaired text to train the E2E SLU system to classify speech using at least one of the one or more sets of semantic labels. The method may include training a text-to-intent model using the set of unpaired text; and training a speech-to-intent model using the text-to-intent model. Alternatively or additionally, the method may include using a text-to-speech (TTS) system to generate synthetic speech from the unpaired text; and training the E2E SLU system using the synthetic speech.
    Type: Application
    Filed: April 7, 2020
    Publication date: October 7, 2021
    Inventors: Hong-Kwang Jeff Kuo, Yinghui Huang, Samuel Thomas, Kartik Audhkhasi, Michael Alan Picheny
  • Publication number: 20210280167
    Abstract: According to one embodiment, a method, computer system, and computer program product for customizing the rendering of a synthesized speech prompt is provided. The present invention may include extracting prosodic information from a received audio recording of a prompt by parsing the text corresponding with the prompt and generating phonetic units, aligning the phonetic units with the audio recording, and calculating, based on the alignment, prosodic values for the phonetic units. The invention may further include adapting the prosodic values to match a text-to-speech voice in use, and then synthesizing speech for the prompt based upon the adapted prosodic information.
    Type: Application
    Filed: March 4, 2020
    Publication date: September 9, 2021
    Inventors: Maria E. Smith, Radek Kazbunda, Michael Alan Picheny, Raul Fernandez
  • Publication number: 20210082437
    Abstract: Aspects of the present disclosure describe techniques for identifying and recovering out-of-vocabulary words in transcripts of a voice data recording using word recognition models and word sub-unit recognition models. An example method generally includes receiving a voice data recording for transcription into a textual representation of the voice data recording. The voice data recording is transcribed into the textual representation using a word recognition model. An unknown word is identified in the textual representation, and the unknown word is reconstructed based on recognition of sub-units of the unknown word generated by a sub-unit recognition model. The textual representation of the voice data recording is modified by replacing the unknown word with the reconstruction of the unknown word, and the modified textual representation is output.
    Type: Application
    Filed: September 13, 2019
    Publication date: March 18, 2021
    Inventors: SAMUEL THOMAS, KARTIK AUDHKHASI, ZOLTAN TUESKE, YINGHUI HUANG, MICHAEL ALAN PICHENY
  • Publication number: 20210065680
    Abstract: In an approach to soft-forgetting training, one or more computer processors train a first model utilizing one or more training batches wherein each training batch of the one or more training batches comprises one or more blocks of information. The one or more computer processors, responsive to a completion of the training of the first model, initiate a training of a second model utilizing the one or more training batches. The one or more computer processors jitter a random block size for each block of information for each of the one or more training batches for the second model. The one or more computer processors unroll the second model over one or more non-overlapping contiguous jittered blocks of information. The one or more computer processors, responsive to the unrolling of the second model, reduce overfitting for the second model by applying twin regularization.
    Type: Application
    Filed: August 27, 2019
    Publication date: March 4, 2021
    Inventors: Kartik Audhkhasi, George Andrei Saon, Zoltan Tueske, Brian E. D. Kingsbury, Michael Alan Picheny
  • Patent number: 8145491
    Abstract: When pitch of a speech segment is being modified from a current pitch to a requested pitch, and the difference between these is relatively large, a pitch modification algorithm is used to modify the pitch of the speech segment. When the difference between current and requested pitches is relatively small, the pitch of the speech segment is not modified. After one or the other speech modification techniques are used, then the resultant modified speech segment is overlapped and added to previously modified speech segments. A modification ratio is determined in order to quantify the difference between the current and requested pitches for a speech segment. The modification ratio is a ratio between the requested and current pitches. Low and high ratio thresholds are used to determine when pitch is being modified to a predetermined high degree, and whether pitch of the speech segment will or will not be modified.
    Type: Grant
    Filed: July 30, 2002
    Date of Patent: March 27, 2012
    Assignee: Nuance Communications, Inc.
    Inventors: Wael Mohamed Hamza, Michael Alan Picheny
  • Patent number: 7996211
    Abstract: A method, apparatus and computer instructions is provided for fast semi-automatic semantic annotation. Given a limited annotated corpus, the present invention assigns a tag and a label to each word of the next limited annotated corpus using a parser engine, a similarity engine, and a SVM engine. A rover then combines the parse trees from the three engines and annotates the next chunk of limited annotated corpus with confidence, such that the efforts required for human annotation is reduced.
    Type: Grant
    Filed: May 20, 2008
    Date of Patent: August 9, 2011
    Assignee: Nuance Communications, Inc.
    Inventors: Yuqing Gao, Michael Alan Picheny, Ruhi Sarikaya
  • Patent number: 7610191
    Abstract: A method, apparatus and computer instructions is provided for fast semi-automatic semantic annotation. Given a limited annotated corpus, the present invention assigns a tag and a label to each word of the next limited annotated corpus using a parser engine, a similarity engine, and a SVM engine. A rover then combines the parse trees from the three engines and annotates the next chunk of limited annotated corpus with confidence, such that the efforts required for human annotation is reduced.
    Type: Grant
    Filed: October 6, 2004
    Date of Patent: October 27, 2009
    Assignee: Nuance Communications, Inc.
    Inventors: Yuqing Gao, Michael Alan Picheny, Ruhi Sarikaya
  • Patent number: 7490042
    Abstract: A technique for producing speech output in an automatic dialog system in accordance with a detected context is provided. Communication is received from a user at the automatic dialog system. A context of the communication from the user is detected in a context detector of the automatic dialog system. A message is created in a natural language generator of the automatic dialog system in communication with the context detector. The message is conveyed to the user through a speech synthesis system of the automatic dialog system, in communication with the natural language generator and the context detector. Responsive to a detected level of ambient noise, the context detector provides at least one command in a markup language to cause the natural language generator to create the message using maximally intelligible words and to cause the speech synthesis system to convey the message with increased volume and decreased speed.
    Type: Grant
    Filed: March 29, 2005
    Date of Patent: February 10, 2009
    Assignee: International Business Machines Corporation
    Inventors: Ellen Marie Eide, Wael Mohamed Hamza, Michael Alan Picheny
  • Publication number: 20080312921
    Abstract: In a speech recognition system, the combination of a log-linear model with a multitude of speech features is provided to recognize unknown speech utterances. The speech recognition system models the posterior probability of linguistic units relevant to speech recognition using a log-linear model. The posterior model captures the probability of the linguistic unit given the observed speech features and the parameters of the posterior model. The posterior model may be determined using the probability of the word sequence hypotheses given a multitude of speech features. Log-linear models are used with features derived from sparse or incomplete data. The speech features that are utilized may include asynchronous, overlapping, and statistically non-independent speech features. Not all features used in training need to appear in testing/recognition.
    Type: Application
    Filed: August 20, 2008
    Publication date: December 18, 2008
    Inventors: Scott E. Axelrod, Sreeram Viswanath Balakrishnan, Stanley F. Chen, Yuging Gao, Rameah A. Gopinath, Hong-Kwang Kuo, Benoit Maison, David Nahamoo, Michael Alan Picheny, George A. Saon, Geoffrey G. Zweig
  • Patent number: 7464031
    Abstract: In a speech recognition system, the combination of a log-linear model with a multitude of speech features is provided to recognize unknown speech utterances. The speech recognition system models the posterior probability of linguistic units relevant to speech recognition using a log-linear model. The posterior model captures the probability of the linguistic unit given the observed speech features and the parameters of the posterior model. The posterior model may be determined using the probability of the word sequence hypotheses given a multitude of speech features. Log-linear models are used with features derived from sparse or incomplete data. The speech features that are utilized may include asynchronous, overlapping, and statistically non-independent speech features. Not all features used in training need to appear in testing/recognition.
    Type: Grant
    Filed: November 28, 2003
    Date of Patent: December 9, 2008
    Assignee: International Business Machines Corporation
    Inventors: Scott E. Axelrod, Sreeram Viswanath Balakrishnan, Stanley F. Chen, Yuging Gao, Ramesh A. Gopinath, Hong-Kwang Kuo, Benoit Maison, David Nahamoo, Michael Alan Picheny, George A. Saon, Geoffrey G. Zweig
  • Publication number: 20080221874
    Abstract: A method, apparatus and computer instructions is provided for fast semi-automatic semantic annotation. Given a limited annotated corpus, the present invention assigns a tag and a label to each word of the next limited annotated corpus using a parser engine, a similarity engine, and a SVM engine. A rover then combines the parse trees from the three engines and annotates the next chunk of limited annotated corpus with confidence, such that the efforts required for human annotation is reduced.
    Type: Application
    Filed: May 20, 2008
    Publication date: September 11, 2008
    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Yuging Cao, Michael Alan Picheny, Ruhi Sarikaya
  • Patent number: 6850888
    Abstract: A method and apparatus are disclosed for training a pattern recognition system, such as a speech recognition system, using an improved objective function. The concept of rank likelihood, previously applied only to the decding process, is applied in a novel manner to the parameter estimation of the training phase of a pattern recognition system. The disclosed objective function is based on a pseudo-rank likelihood that not only maximizes the likelihood of an observation for the correct class, but also minimizes the likelihoods of the observation for all other classes, such that the discrimination between classes is maximized. A training process is disclosed that utilizes the pseudo-rank likelihood objective function to identify model parameters that will result in a pattern recognizer with the lowest possible recognition error rate. The discrete nature of the rank-based rank likelihood objective function is transformed to allow the parameter estimations to be optimized during the training phase.
    Type: Grant
    Filed: October 6, 2000
    Date of Patent: February 1, 2005
    Assignee: International Business Machines Corporation
    Inventors: Yuqing Gao, Yongxin Li, Michael Alan Picheny
  • Patent number: 6754625
    Abstract: There is provided a method for augmenting an alternate word list generated by a speech recognition system. The alternate word list includes at least one potentially correct word for replacing a wrongly decoded word. The method includes the step of identifying at least one acoustically confusable word with respect to the wrongly decoded word. The alternate word list is augmented with the at least one acoustically confusable word.
    Type: Grant
    Filed: December 26, 2000
    Date of Patent: June 22, 2004
    Assignee: International Business Machines Corporation
    Inventors: Peder Andreas Olsen, Michael Alan Picheny, Harry W. Printz, Karthik Visweswariah
  • Publication number: 20040024600
    Abstract: When pitch of a speech segment is being modified from a current pitch to a requested pitch, and the difference between these is relatively large, a pitch modification algorithm is used to modify the pitch of the speech segment. When the difference between current and requested pitches is relatively small, the pitch of the speech segment is not modified. After one or the other speech modification techniques are used, then the resultant modified speech segment is overlapped and added to previously modified speech segments. A modification ratio is determined in order to quantify the difference between the current and requested pitches for a speech segment. The modification ratio is a ratio between the requested and current pitches. Low and high ratio thresholds are used to determine when pitch is being modified to a predetermined high degree, and whether pitch of the speech segment will or will not be modified.
    Type: Application
    Filed: July 30, 2002
    Publication date: February 5, 2004
    Applicant: International Business Machines Corporation
    Inventors: Wael Mohamed Hamza, Michael Alan Picheny
  • Publication number: 20020116191
    Abstract: There is provided a method for augmenting an alternate word list generated by a speech recognition system. The alternate word list includes at least one potentially correct word for replacing a wrongly decoded word. The method includes the step of identifying at least one acoustically confusable word with respect to the wrongly decoded word. The alternate word list is augmented with the at least one acoustically confusable word.
    Type: Application
    Filed: December 26, 2000
    Publication date: August 22, 2002
    Applicant: International Business Machines Corporation
    Inventors: Peder Andreas Olsen, Michael Alan Picheny, Harry W. Printz, Karthik Visweswariah
  • Patent number: 6073096
    Abstract: A method of speech recognition, in accordance with the present invention includes the steps of grouping acoustics to form classes based on acoustic features, clustering training speakers by the classes to provide class-specific cluster systems, selecting from the cluster systems, a subset of cluster systems closest to adaptation data from a test speaker, transforming the subset of cluster systems to bring the subset of cluster systems closer to the test speaker based on the adaptation data to form adapted cluster systems and combining the adapted cluster systems to create a speaker adapted system for decoding speech from the test speaker. System and methods for building speech recognition systems as well as adapting speaker systems for class-specific speaker clusters are included.
    Type: Grant
    Filed: February 4, 1998
    Date of Patent: June 6, 2000
    Assignee: International Business Machines Corporation
    Inventors: Yuqing Gao, Mukund Padmanabhan, Michael Alan Picheny
  • Patent number: 6023673
    Abstract: A speech coding apparatus and method uses a hierarchy of prototype sets to code an utterance while consuming fewer computing resources. The value of at least one feature of an utterance is measured during each of a series of successive time intervals to produce a series of feature vector signals representing the feature values. A plurality of level subsets of prototype vector signals is computed, wherein each prototype vector signal in a higher level subset is associated with at least one prototype vector signal in a lower level subset. Each level subset contains a plurality of prototype vector signals, with lower level subsets containing more prototypes than higher level subsets. The closeness of the feature value of the first feature vector signal is compared to the parameter values of prototype vector signals in the first level subset of prototype vector signals to obtain a ranked list of prototype match scores for the first feature vector signal and each prototype vector signal in the first level subset.
    Type: Grant
    Filed: June 4, 1997
    Date of Patent: February 8, 2000
    Assignee: International Business Machines Corporation
    Inventors: Raimo Bakis, David Nahamoo, Michael Alan Picheny, Jan Sedivy
  • Patent number: 5995590
    Abstract: A method and apparatus is disclosed that allows people to carry on unobtrusive phone conversations in business or other settings where it is either not possible or impolite to talk. In the system of FIG. 1, the telephone user one will listen in the same manner as with a regular telephone. However, he will not speak into the telephone microphone. User one instead employs a unit including a keyboard to enter the text corresponding to what he wants to say. The text is converted into a synthesized speech using TTS apparatus and a voice output is sent to the microphone of the phone apparatus. The telephone apparatus transmits the synthesized voice signal over a standard telephone line to a unit including a conventional telephone speaker 26 and telephone microphone. User two, the party using the telephone at the other end, listens to a synthesized voice, but user one listens to the actual voice of user two with the telephone speaker, unless user two is also using a system similar to that of user one.
    Type: Grant
    Filed: March 5, 1998
    Date of Patent: November 30, 1999
    Assignee: International Business Machines Corporation
    Inventors: Peter Thomas Brunet, Abraham P. Ittycheriah, Chandrasekhar Narayanaswami, Michael Alan Picheny, Bhuvana Ramabhadran