Patents by Inventor Juergen Fritsch
Juergen Fritsch 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: 9489433Abstract: A dataset is searched using inclusion set criteria to produce an inclusion set and exclusion set criteria to produce an exclusion set. A set of unique content elements is identified from the inclusion set and the exclusion set. Metrics are derived from the inclusion set, exclusion set, and set of unique content elements, such as a measure, for each unique content element, of the absolute value of the difference between the percentage of records in the inclusion set containing the unique content element and the percentage of records in the exclusion set containing the unique content element. The unique content element set may be sorted and displayed in decreasing order of the above-referenced absolute value. The content element set may be filtered. Individual content elements may be excluded from the content set. A predictive model may be generated based on the resulting version of the content element set.Type: GrantFiled: January 27, 2016Date of Patent: November 8, 2016Assignee: MModal IP LLCInventors: Jonathan A. Handler, Juergen Fritsch
-
Patent number: 9454965Abstract: Techniques are disclosed for facilitating the process of proofreading draft transcripts of spoken audio streams. In general, proofreading of a draft transcript is facilitated by playing back the corresponding spoken audio stream with an emphasis on those regions in the audio stream that are highly relevant or likely to have been transcribed incorrectly. Regions may be emphasized by, for example, playing them back more slowly than regions that are of low relevance and likely to have been transcribed correctly. Emphasizing those regions of the audio stream that are most important to transcribe correctly and those regions that are most likely to have been transcribed incorrectly increases the likelihood that the proofreader will accurately correct any errors in those regions, thereby improving the overall accuracy of the transcript.Type: GrantFiled: September 11, 2015Date of Patent: September 27, 2016Assignee: MModal IP LLCInventors: Kjell Schubert, Juergen Fritsch, Michael Finke, Detlef Koll
-
Publication number: 20160196821Abstract: A system is provided for training an acoustic model for use in speech recognition. In particular, such a system may be used to perform training based on a spoken audio stream and a non-literal transcript of the spoken audio stream. Such a system may identify text in the non-literal transcript which represents concepts having multiple spoken forms. The system may attempt to identify the actual spoken form in the audio stream which produced the corresponding text in the non-literal transcript, and thereby produce a revised transcript which more accurately represents the spoken audio stream. The revised, and more accurate, transcript may be used to train the acoustic model, thereby producing a better acoustic model than that which would be produced using conventional techniques, which perform training based directly on the original non-literal transcript.Type: ApplicationFiled: March 10, 2016Publication date: July 7, 2016Applicant: MModal IP LLCInventors: Girija Yegnanarayanan, Michael Finke, Juergen Fritsch, Detlef Koll, Monika Woszczyna
-
Publication number: 20160179770Abstract: An automatic speech recognizer is used to produce a structured document representing the contents of human speech. A best practice is applied to the structured document to produce a conclusion, such as a conclusion that required information is missing from the structured document. Content is inserted into the structured document based on the conclusion, thereby producing a modified document. The inserted content may be obtained by prompting a human user for the content and receiving input representing the content from the human user.Type: ApplicationFiled: February 26, 2016Publication date: June 23, 2016Applicant: MModal IP LLCInventors: Detlef Koll, Juergen Fritsch, Michael Finke
-
Publication number: 20160140193Abstract: A dataset is searched using inclusion set criteria to produce an inclusion set and exclusion set criteria to produce an exclusion set. A set of unique content elements is identified from the inclusion set and the exclusion set. Metrics are derived from the inclusion set, exclusion set, and set of unique content elements, such as a measure, for each unique content element, of the absolute value of the difference between the percentage of records in the inclusion set containing the unique content element and the percentage of records in the exclusion set containing the unique content element. The unique content element set may be sorted and displayed in decreasing order of the above-referenced absolute value. The content element set may be filtered. Individual content elements may be excluded from the content set. A predictive model may be generated based on the resulting version of the content element set.Type: ApplicationFiled: January 27, 2016Publication date: May 19, 2016Applicant: MModal IP LLCInventors: Jonathan A. Handler, Juergen Fritsch
-
Publication number: 20160078861Abstract: A system is provided for training an acoustic model for use in speech recognition. In particular, such a system may be used to perform training based on a spoken audio stream and a non-literal transcript of the spoken audio stream. Such a system may identify text in the non-literal transcript which represents concepts having multiple spoken forms. The system may attempt to identify the actual spoken form in the audio stream which produced the corresponding text in the non-literal transcript, and thereby produce a revised transcript which more accurately represents the spoken audio stream. The revised, and more accurate, transcript may be used to train the acoustic model using discriminative training techniques, thereby producing a better acoustic model than that which would be produced using conventional techniques, which perform training based directly on the original non-literal transcript.Type: ApplicationFiled: November 16, 2015Publication date: March 17, 2016Applicant: MMODAL IP LLCInventors: Lambert Mathias, Girija Yegnanarayanan, Juergen Fritsch
-
Patent number: 9286896Abstract: A system is provided for training an acoustic model for use in speech recognition. In particular, such a system may be used to perform training based on a spoken audio stream and a non-literal transcript of the spoken audio stream. Such a system may identify text in the non-literal transcript which represents concepts having multiple spoken forms. The system may attempt to identify the actual spoken form in the audio stream which produced the corresponding text in the non-literal transcript, and thereby produce a revised transcript which more accurately represents the spoken audio stream. The revised, and more accurate, transcript may be used to train the acoustic model, thereby producing a better acoustic model than that which would be produced using conventional techniques, which perform training based directly on the original non-literal transcript.Type: GrantFiled: May 16, 2014Date of Patent: March 15, 2016Assignee: MModal IP LLCInventors: Girija Yegnanarayanan, Michael Finke, Juergen Fritsch, Detlef Koll, Monika Woszczyna
-
Patent number: 9275643Abstract: An automatic speech recognizer is used to produce a structured document representing the contents of human speech. A best practice is applied to the structured document to produce a conclusion, such as a conclusion that required information is missing from the structured document. Content is inserted into the structured document based on the conclusion, thereby producing a modified document. The inserted content may be obtained by prompting a human user for the content and receiving input representing the content from the human user.Type: GrantFiled: July 9, 2014Date of Patent: March 1, 2016Assignee: MModal IP LLCInventors: Detlef Koll, Juergen Fritsch, Michael Finke
-
Patent number: 9251203Abstract: A dataset is searched using inclusion set criteria to produce an inclusion set and exclusion set criteria to produce an exclusion set. A set of unique content elements is identified from the inclusion set and the exclusion set. Metrics are derived from the inclusion set, exclusion set, and set of unique content elements, such as a measure, for each unique content element, of the absolute value of the difference between the percentage of records in the inclusion set containing the unique content element and the percentage of records in the exclusion set containing the unique content element. The unique content element set may be sorted and displayed in decreasing order of the above-referenced absolute value. The content element set may be filtered. Individual content elements may be excluded from the content set. A predictive model may be generated based on the resulting version of the content element set.Type: GrantFiled: December 20, 2013Date of Patent: February 2, 2016Assignee: MModal IP LLCInventors: Jonathan A. Handler, Juergen Fritsch
-
Publication number: 20160005402Abstract: Techniques are disclosed for facilitating the process of proofreading draft transcripts of spoken audio streams. In general, proofreading of a draft transcript is facilitated by playing back the corresponding spoken audio stream with an emphasis on those regions in the audio stream that are highly relevant or likely to have been transcribed incorrectly. Regions may be emphasized by, for example, playing them back more slowly than regions that are of low relevance and likely to have been transcribed correctly. Emphasizing those regions of the audio stream that are most important to transcribe correctly and those regions that are most likely to have been transcribed incorrectly increases the likelihood that the proofreader will accurately correct any errors in those regions, thereby improving the overall accuracy of the transcript.Type: ApplicationFiled: September 11, 2015Publication date: January 7, 2016Applicant: MModal IP LLCInventors: Kjell Schubert, Juergen Fritsch, Michael Finke, Detlef Koll
-
Patent number: 9190050Abstract: A system is provided for training an acoustic model for use in speech recognition. In particular, such a system may be used to perform training based on a spoken audio stream and a non-literal transcript of the spoken audio stream. Such a system may identify text in the non-literal transcript which represents concepts having multiple spoken forms. The system may attempt to identify the actual spoken form in the audio stream which produced the corresponding text in the non-literal transcript, and thereby produce a revised transcript which more accurately represents the spoken audio stream. The revised, and more accurate, transcript may be used to train the acoustic model using discriminative training techniques, thereby producing a better acoustic model than that which would be produced using conventional techniques, which perform training based directly on the original non-literal transcript.Type: GrantFiled: April 3, 2014Date of Patent: November 17, 2015Assignee: MModal IP LLCInventors: Lambert Mathias, Girija Yegnanarayanan, Juergen Fritsch
-
Patent number: 9135917Abstract: Techniques are disclosed for facilitating the process of proofreading draft transcripts of spoken audio streams. In general, proofreading of a draft transcript is facilitated by playing back the corresponding spoken audio stream with an emphasis on those regions in the audio stream that are highly relevant or likely to have been transcribed incorrectly. Regions may be emphasized by, for example, playing them back more slowly than regions that are of low relevance and likely to have been transcribed correctly. Emphasizing those regions of the audio stream that are most important to transcribe correctly and those regions that are most likely to have been transcribed incorrectly increases the likelihood that the proofreader will accurately correct any errors in those regions, thereby improving the overall accuracy of the transcript.Type: GrantFiled: June 27, 2014Date of Patent: September 15, 2015Assignee: MModal IP LLCInventors: Kjell Schubert, Juergen Fritsch, Michael Finke, Detlef Koll
-
Publication number: 20150154168Abstract: A system includes a document corpus containing structured documents, which contain both text and annotations of the text. The system also includes a search engine which is adapted to perform structured searches of the structured documents. As new types of annotations are added to the system, the search engine is updated automatically to become capable of performing structured searches for the new types of annotations. For example, if a new natural language processing (NLP) component, adapted to generate annotations of a new type, is added to the system, then the system automatically updates a query language to include a definition of the new type of annotation. The search engine may then immediately be capable of processing structured queries which refer to the new type of annotation.Type: ApplicationFiled: February 3, 2015Publication date: June 4, 2015Applicant: MMODAL IP LLCInventors: Detlef Koll, Juergen Fritsch
-
Publication number: 20150095025Abstract: Non-verbalized tokens, such as punctuation, are automatically predicted and inserted into a transcription of speech in which the tokens were not explicitly verbalized. Token prediction may be integrated with speech decoding, rather than performed as a post-process to speech decoding.Type: ApplicationFiled: December 16, 2014Publication date: April 2, 2015Applicant: Multimodal Technologies, LLCInventors: Juergen Fritsch, Anoop Deoras, Detlef Koll
-
Patent number: 8959102Abstract: A system includes a document corpus containing structured documents, which contain both text and annotations of the text. The system also includes a search engine which is adapted to perform structured searches of the structured documents. As new types of annotations are added to the system, the search engine is updated automatically to become capable of performing structured searches for the new types of annotations. For example, if a new natural language processing (NLP) component, adapted to generate annotations of a new type, is added to the system, then the system automatically updates a query language to include a definition of the new type of annotation. The search engine may then immediately be capable of processing structured queries which refer to the new type of annotation.Type: GrantFiled: October 8, 2011Date of Patent: February 17, 2015Assignee: MModal IP LLCInventors: Detlef Koll, Juergen Fritsch
-
Patent number: 8918317Abstract: Non-verbalized tokens, such as punctuation, are automatically predicted and inserted into a transcription of speech in which the tokens were not explicitly verbalized. Token prediction may be integrated with speech decoding, rather than performed as a post-process to speech decoding.Type: GrantFiled: September 25, 2009Date of Patent: December 23, 2014Assignee: Multimodal Technologies, LLCInventors: Juergen Fritsch, Anoop Deoras, Detlef Koll
-
Publication number: 20140343963Abstract: A computer system generates an initial set of billing codes based on one or more documents (e.g., clinical notes) representing a patient encounter, such as clinical notes created by a physician. The system expands the initial set of billing codes based on a billing code standard to create an expanded set of billing codes for consideration by the physician. The system provides output representing the expanded billing code set to the physician. The physician selects one or more billing codes from the expanded billing code set for inclusion in a final billing code set for use in a bill for the services provided in the patient encounter.Type: ApplicationFiled: March 18, 2014Publication date: November 20, 2014Applicant: MModal IP LLCInventors: Juergen Fritsch, Vasudevan Jagannathan
-
Publication number: 20140343939Abstract: A system is provided for training an acoustic model for use in speech recognition. In particular, such a system may be used to perform training based on a spoken audio stream and a non-literal transcript of the spoken audio stream. Such a system may identify text in the non-literal transcript which represents concepts having multiple spoken forms. The system may attempt to identify the actual spoken form in the audio stream which produced the corresponding text in the non-literal transcript, and thereby produce a revised transcript which more accurately represents the spoken audio stream. The revised, and more accurate, transcript may be used to train the acoustic model using discriminative training techniques, thereby producing a better acoustic model than that which would be produced using conventional techniques, which perform training based directly on the original non-literal transcript.Type: ApplicationFiled: April 3, 2014Publication date: November 20, 2014Applicant: MModal IP LLCInventors: Lambert Mathias, Girija Yegnanarayanan, Juergen Fritsch
-
Publication number: 20140324423Abstract: An automatic speech recognizer is used to produce a structured document representing the contents of human speech. A best practice is applied to the structured document to produce a conclusion, such as a conclusion that required information is missing from the structured document. Content is inserted into the structured document based on the conclusion, thereby producing a modified document. The inserted content may be obtained by prompting a human user for the content and receiving input representing the content from the human user.Type: ApplicationFiled: July 9, 2014Publication date: October 30, 2014Inventors: Detlef Koll, Juergen Fritsch, Michael Finke
-
Publication number: 20140309995Abstract: Techniques are disclosed for facilitating the process of proofreading draft transcripts of spoken audio streams. In general, proofreading of a draft transcript is facilitated by playing back the corresponding spoken audio stream with an emphasis on those regions in the audio stream that are highly relevant or likely to have been transcribed incorrectly. Regions may be emphasized by, for example, playing them back more slowly than regions that are of low relevance and likely to have been transcribed correctly. Emphasizing those regions of the audio stream that are most important to transcribe correctly and those regions that are most likely to have been transcribed incorrectly increases the likelihood that the proofreader will accurately correct any errors in those regions, thereby improving the overall accuracy of the transcript.Type: ApplicationFiled: June 27, 2014Publication date: October 16, 2014Inventors: Kjell Schubert, Juergen Fritsch, Michael Finke, Detlef Koll