Abstract: In one aspect, a method for processing media includes accepting a query. One or more language patterns are identified that are similar to the query. A putative instance of the query is located in the media. The putative instance is associated with a corresponding location in the media. The media in a vicinity of the putative instance is compared to the identified language patterns and data characterizing the putative instance of the query is provided according to the comparing of the media to the language patterns, for example, as a score for the putative instance that is determined according to the comparing of the media to the language patterns.

Abstract: An approach to words spotting processes a query including a sequence of terms (e.g., words) to identify one or more subsequences that constitute segments (e.g., phrases) that are likely to occur spoken together in the audio begin searched. The segments are searched for as units. An advantage can include improved accuracy as compared to searching for the terms individually.

Abstract: Systems, method, and apparatus for processing a speech utterance or audio record that includes receiving one or more feature vectors characterizing the speech utterance or audio record, each feature vector having a plurality of feature elements, each feature element being associated with a spectral representation of a characteristic of one of a plurality of sequential segments of the speech utterance or audio record; and processing the one or more feature vectors in a rank order filter to obtain one or more normalized feature vectors, each normalized feature vector having a plurality of normalized feature elements corresponding to the plurality of feature elements.

Abstract: Some general aspects of the invention relate to systems and computer-implemented methods of generating a treemap display. A collection of data elements characterized by a first attribute is accepted, and some data elements are grouped into a first set of data elements according to a first rule associated with the first attribute. A treemap field is partitioned into a collection of cells according to the grouping result, and the collection of cells includes a first cell representing the first set of data elements. The first cell has a first dimension corresponding to a value of the first attribute of the first set of data elements. The first set of data elements is then divided into a collection of subsets of data elements according to a second rule. Correspondingly, the first cell of the treemap field is partitioned into a collection of sub-cells according to the division. Each sub-cell represents a respective one of the plurality of subsets of data elements.

Abstract: An approach to scoring acoustically-based events, such as hypothesized instances of keywords, in a speech processing system make use of scores of individual components of the event. Data characterizing an instance of an event are first accepted. This data includes a score for the event. The event is associated with a number of component events from a set of component events, such as a set of phonemes. Probability models are also accepted for component scores associated with each of the set of component events in each of two of more possible classes of the event, such as a class of true occurrences of the event and a class of false detections of the event. The event is then scored. This scoring includes computing a probability of one of the two or more possible classes for the event using the accepted probability models.

Abstract: An approach to improving the performance of a wordspotting system includes providing an interface for interactive improvement of a phonetic representation of a query based on an operator identifying true detections and false alarms in a data set.

Abstract: This invention relates to processing of audio files, and more specifically, to an improved technique of searching audio. More particularly, a method and system for processing audio using a multi-stage searching process is disclosed.

Abstract: An approach to alignment of transcripts with recorded audio is tolerant of moderate transcript inaccuracies, untranscribed speech, and significant non-speech noise. In one aspect, a number of search terms are formed from the transcript such that each search term is associated with a location within the transcript. Possible locations of the search terms are then determined in the audio recording. The audio recording and the transcript are then aligned using the possible locations of the search terms. In another aspect a search expression is accepted, and then a search is performed for spoken occurrences of the search expression in an audio recording. This search includes searching for text occurrences of the search expression in a text transcript of the audio recording, and searching for spoken occurrences of the search expression in the audio recording.

Abstract: In some aspects, a wordspotter is used to locate occurrences in an audio corpus of each of a set of predetermined subword units, which may be phoneme sequences. To locate a query (e.g., a keyword or phrase) in the audio corpus, constituent subword units in the query are indentified and then locations of those subwords are determined based on the locations of those subword units determined earlier by the wordspotter, for example, using a pre-built inverted index that maps subword units to their locations.

Abstract: An information retrieval system for searching a corpus is configured to operate in a manner that optimizes the consistency of a user experience given a subset of a corpus and a search query.

Abstract: A wordspotting system is applied to a speech source in a preliminary processing phase. The putative hits corresponding to queries (e.g., keywords, key phrases, or more complex queries that may include Boolean expressions and proximity operators) are used to control a speech recognizer. The control can include one or more of application of a time specification that is determined from the putative hits for selecting an interval of the speech source to which to apply the speech recognizer; application of a grammar specification determined from the putative hits that is used by the speech recognizer, and application of a specification of a lattice or pruning specification that is used by the recognizer to limit or guide the recognizer in recognition of the speech source.

Abstract: An approach to alignment of transcripts with recorded audio is tolerant of moderate transcript inaccuracies, untranscribed speech, and significant non-speech noise. In one aspect, a number of search terms are formed from the transcript such that each search term is associated with a location within the transcript. Possible locations of the search terms are then determined in the audio recording. The audio recording and the transcript are then aligned using the possible locations of the search terms. In another aspect a search expression is accepted, and then a search is performed for spoken occurrences of the search expression in an audio recording. This search includes searching for text occurrences of the search expression in a text transcript of the audio recording, and searching for spoken occurrences of the search expression in the audio recording.

Abstract: An approach to evaluating a person's speech skills includes automatically processing speech of a person and text some or all of which corresponds to the speech. In some examples, a job application procedure includes collecting speech from an applicant, and using text corresponding to the collected speech to automatically assess speech skills of the applicant. The text may include text that is presented to the applicant and the speech collected from the applicant can include the applicant reading the presented text.

Abstract: An approach to accessing audio or multimedia content uses associated text sources to segment the content and/or to locate entities in the content. A user interface then provides a user with a way to navigate the content in a non-linear manner based on the segmentation or linking of text entities with locations in the content. The user interface can also provide a way to edit segment-specific content and to publish individual segments of the content. The output of the system, for instance the individual segments of annotated content, can be used to syndicate and/or to improve discoverability of the content.

Abstract: An approach to alignment of transcripts with recorded audio is tolerant of moderate transcript inaccuracies, untranscribed speech, and significant non-speech noise. In one aspect, a number of search terms are formed from the transcript such that each search term is associated with a location within the transcript. Possible locations of the search terms are then determined in the audio recording. The audio recording and the transcript are then aligned using the possible locations of the search terms. In another aspect a search expression is accepted, and then a search is performed for spoken occurrences of the search expression in an audio recording. This search includes searching for text occurrences of the search expression in a text transcript of the audio recording, and searching for spoken occurrences of the search expression in the audio recording.

Abstract: An approach to wordspotting (180) using query data from one or more spoken instance of a query (140). The query data is processed to determining a representation of the query (160) that defines multiple sequences of subword (130) units each representing the query. Then putative instances of the query (190) are located in input data from an audio signal using the determined representation of the query.