Nuance Communications, Inc.

Intel Corporation

International Business Machines Corporation

Jonathan Mamou

Moshe Wasserblat

Oren Pereg

Michel Assayag

Bhuvana Ramabhadran

Paul Joseph Vozila

Ron Hoory

Abhinav Sethy

Nathan M Bodenstab

Oren Shamir

Itzhack Goldberg

Shmuel Ben-Yehuda

Yosi Mass

Ohad Greenshpan

Benjamin Sznajder

Technologies for detecting an end of a sentence in automatic speech recognition are disclosed. An automatic speech recognition device may acquire speech data, and identify phonemes and words of the speech data. The automatic speech recognition device may perform a syntactic parse based on the recognized words, and determine an end of a sentence based on the syntactic parse. For example, if the syntactic parse indicates that a certain set of consecutive recognized words form a syntactically complete and correct sentence, the automatic speech recognition device may determine that there is an end of a sentence at the end of that set of words.

Technologies for end-of-sentence detection using syntactic coherence

A system and method for syntactic re-ranking of possible transcriptions generated by automatic speech recognition are disclosed. A computer system accesses acoustic data for a recorded spoken language and generates a plurality of potential transcriptions for the acoustic data. The computer system scores the plurality of potential transcriptions to create an initial likelihood score for the plurality of potential transcriptions. For a particular potential transcription in the plurality of transcriptions, the computer system generates a syntactical likelihood score. The computer system creates an adjusted score for the particular potential transcription by combining the initial likelihood score and the syntactic likelihood score for the particular potential transcription.

Syntactic re-ranking of potential transcriptions during automatic speech recognition

A method and system are provided of spoken document retrieval using multiple search transcription indices. The method includes receiving a query input formed of one or more query terms and determining a type of a query term, wherein a type includes a term in a speech recognition vocabulary or a term not in a speech recognition vocabulary. One or more indices of search transcriptions are selected for searching the query term based on the type of the query term. The one or more indices are generated using different speech transcription methods. The results for the query term are scored by the one or more indices and the results of the one or more indices for the query term are merged. The results of the one or more query terms are then merged to provide the results for the query.

Spoken document retrieval using multiple speech transcription indices

Techniques disclosed herein include systems and methods for voice-enabled searching. Techniques include a co-occurrence based approach to improve accuracy of the 1-best hypothesis for non-phrase voice queries, as well as for phrased voice queries. A co-occurrence model is used in addition to a statistical natural language model and acoustic model to recognize spoken queries, such as spoken queries for searching a search engine. Given an utterance and an associated list of automated speech recognition n-best hypotheses, the system rescores the different hypotheses using co-occurrence information. For each hypothesis, the system estimates a frequency of co-occurrence within web documents. Combined scores from a speech recognizer and a co-occurrence engine can be combined to select a best hypothesis with a lower word error rate.

Accuracy improvement of spoken queries transcription using co-occurrence information

A method and system of indexing speech data. The method includes indexing word transcripts including a timestamp for a word occurrence; and indexing sub-word transcripts including a timestamp for a sub-word occurrence. A timestamp in the index indicates the time and duration of occurrence of the word or sub-word in the speech data, and word and sub-word occurrences can be correlated using the timestamps. A method of searching speech transcripts is also provided in which a search query in the form of a phrase to be searched includes at least one in-vocabulary word and at least one out-of-vocabulary word. The method of searching includes extracting the search terms from the phrase, retrieving a list of occurrence of words for an in-vocabulary search term from an index of words having timestamps, retrieving a list of occurrences of sub-words for an out-of-vocabulary search term from an index of sub-words having timestamps, and merging the retrieved lists of occurrences of words and sub-words according to their timestamps.

Method and system of indexing speech data

A method for information retrieval includes extracting from a video document visual data items and textual data items that occur in the document at respective occurrence times. Indexing records, which index both the visual and the textual data items by their respective occurrence times, are constructed and stored in a memory.

Unified inverted index for video passage retrieval

Techniques include customizing a context structure such as a context menu based on content associated with an object such as a computer file. For example, a computer-implemented method for customizing a context structure associated with an object comprises the following steps. A content analysis is performed on the object. The context structure associated with the object is modified based on at least a portion of results of the content analysis.

Growing community of inventors

Jerusalem, Israel

Jonathan Mamou