Abstract: A method of adapting a speech recognition system to one or more acoustic conditions comprises the steps of: (i) computing cumulative distribution functions based on dimensions of speech vectors associated with training speech data provided to the speech recognition system; (ii) computing cumulative distribution functions based on dimensions of speech vectors associated with test speech data provided to the speech recognition system; (iii) computing a nonlinear transformation mapping based on the cumulative distribution functions associated with the training speech data and the cumulative distribution functions associated with the test speech data; and (iv) applying the nonlinear transformation mapping to speech vectors associated with the test speech data prior to recognition, wherein the speech vectors transformed in accordance with the nonlinear transformation mapping are substantially similar to speech vectors associated with the training speech data.

Abstract: An interactive voice response (IVR) system comprises many different application objects which combine to deliver a voice application. These objects typically include IVR programs or scripts, prompts or voice segments and server scripts or programs for communicating with external systems or databases. Large voice applications can contain hundreds of application objects and potentially thousands of voice segments. The grouping of the application objects becomes more important in the later stages of the process. There is described a method of grouping data objects having different data categories, such as IVR programs or scripts, prompts or voice segments, in an application processing system.

Abstract: The system of the present invention includes a form design component, a form description repository, and a forms processing component Each form used with the system has a layout including a form identifier field with a common location space for each given form of the plurality of different types of forms.

Abstract: An electronic dictionary having an idiom processing function which can automatically identify idioms included in a present sentence from text of a first language, and which can output corresponding translated expressions in a second language. The electronic dictionary is operative to perform a technique which comprises an idiom processing operation which makes automatic identification possible by text capturing, sentence segmenting, local parsing and transfer lexicon matching. The electronic dictionary provides intelligent translation at the idiom level.

Abstract: A method of binarization used in an OCR system involves in determining text pixels by checking, for each pixel, that the difference between its value and the values of a plurality of pixels located at a predetermined distance therefrom is greater than a relative threshold corresponding to the difference in intensities between the text and the background of the image, subsampling the image at a rate corresponding to at least two pixels in order to detect kernels of text, and then binarizing the image pixels only in tiles of several stroke width sides containing text kernels by using in each tile, an absolute threshold estimated in that tile.

Abstract: Methods and apparatus for use in signal timing analysis with respect to a circuit having at least one gate are provided. In one aspect, the invention includes the step of determining a first constraint slew sensitivity value and a second constraint slew sensitivity value for the at least one gate according to a specified bounding technique. Then, a representative signal for the gate is computed in accordance with the first and second values including an arrival time and slew rate, wherein the representative signal bounds signal paths by bounding a maximum slew sensitivity path and a minimum slew sensitivity path. Such a representative signal may be computed for a worst case late-mode analysis and/or a best case early-mode analysis. The bounding technique may be selected by a user at the time the user inputs the schematic of the circuit on which timing analysis is to be performed.

Abstract: A driver circuit having a minimized and/or controllable output common mode voltage comprises a differential amplifier having, a passive element as a biasing source for establishing a bias current in the differential amplifier and a control amplifier operatively coupled to the differential amplifier in a feedback arrangement, the control amplifier generating a control signal. The differential amplifier is responsive to the control signal for providing a voltage at an output of the driver circuit that is substantially independent of an input signal presented to an input of the driver circuit. By eliminating the need for an active device (e.g., transistor) as a bias current source, the output common mode voltage of the driver circuit is minimized. A reference signal coupled to the control amplifier, in conjunction with the feedback arrangement, substantially fixes the output common mode voltage of the driver circuit to a predetermined value.

Abstract: A method and apparatus are disclosed for identifying speakers participating in an audio-video source, whether or not such speakers have been previously registered or enrolled. The speaker identification system uses an enrolled speaker database that includes background models for unenrolled speakers, such as “unenrolled male” or “unenrolled female,” to assign a speaker label to each identified segment. Speaker labels are identified for each speech segment by comparing the segment utterances to the enrolled speaker database and finding the “closest” speaker, if any. A speech segment having an unknown speaker is initially assigned a general speaker label from the set of background models. The “unenrolled” segment is assigned a segment number and receives a cluster identifier assigned by the clustering system.

Abstract: A method of performing speaker adaptation of acoustic models in a band-quantized speech recognition system, wherein the system including one or more acoustic models represented by a feature space of multi-dimensional gaussians, whose dimensions are partitioned into bands, and the gaussian means and covariances within each band are quantized into atoms, comprises the following steps. A decoded segment of a speech signal associated with a particular speaker is obtained. Then, at least one adaptation mapping based on the decoded segment is computed. Lastly, the at least one adaptation mapping is applied to the atoms of the acoustic models to generate one or more acoustic models adapted to the particular speaker. Accordingly, a fast speaker adaptation methodology is provided for use in real-time applications.

Abstract: A method and apparatus are disclosed for automatically transcribing audio information from an audio-video source and concurrently identifying the speakers. The disclosed audio transcription and speaker classification system includes a speech recognition system, a speaker segmentation system and a speaker identification system. A common front-end processor computes feature vectors that are processed along parallel branches in a multi-threaded environment by the speech recognition system, speaker segmentation system and speaker identification system, for example, using a shared memory architecture that acts in a server-like manner to distribute the computed feature vectors to a channel associated with each parallel branch. The speech recognition system produces transcripts with time-alignments for each word in the transcript. The speaker segmentation system separates the speakers and identifies all possible frames where there is a segment boundary between non-homogeneous speech portions.

Abstract: A method of validating production of a biometric attribute allegedly associated with a user comprises the following steps. A first signal is generated representing data associated with the biometric attribute allegedly received in association with the user. A second signal is also generated representing data associated with at least one feature detected in association with the production of the biometric attribute allegedly received from the user. Then, the first signal and the second signal are compared to determine a correlation level between the biometric attribute and the production feature, wherein the validation of the production of the biometric attribute depends on the correlation level. Accordingly, the invention serves to provide substantial assurance that the biometric attribute offered by the user has been physically generated by the user.

Abstract: A phonetic spell checker comprises a dictionary table (10) having a plurality of entries each including an orthography and an associated pronunciation, said pronunciation comprising one or more phonemes; and a weightings table (14) having a plurality of entries, each including a cluster comprising one or more letters, a cluster pronunciation comprising one or more phonemes and a weighting for said pronunciation of said cluster. A user interface receives a word to be checked, a clustering mechanism divides the word into a plurality of clusters, each cluster having one or more pronunciations, each pronunciation comprising one or more phonemes. Pronunciations of the word are ordered according to the associated weightings of the cluster pronunciations in the weighting table; and the dictionary is searched for an orthography whose associated pronunciation matches at least the most heavily weighted pronunciation.

Abstract: An encoding circuit for use with a comparator, includes a plurality of logic elements for receiving an input from a comparator, and a Gray code encoder for receiving an output from the plurality of logic elements. Both first and second type comparator errors (e.g., meta-stability errors and bubble-errors) are substantially eliminated simultaneously by the logic elements.

Abstract: A method of forming an augmented textual training corpus with compound words for use with an associated with a speech recognition system includes computing a measure for a consecutive word pair in the training corpus. The measure is then compared to a threshold value. The consecutive word pair is replaced in the training corpus with a corresponding compound word depending on the result of the comparison between the measure and the threshold value. One or more measures may be employed. A first measure is an average of a direct bigram probability value and a reverse bigram probability value. A second measure is based on mutual information between the words in the pair. A third measure is based on a comparison of the number of times a co-articulated baseform for the pair is preferred over a concatenation of non-co-articulated individual baseforms of the words forming the pair.

Abstract: Disclosed is a method and an apparatus, as well as a computer program, for associating digitizer tablet stroke samples with entries in an electronic personal information management (PIM) tool, such as an electronic organizer or an electronic calendar. A stroke database has a plurality of time stamped stroke entries, and a PIM database has a plurality of entries each having a time associated therewith. A controller searches at least one of the stroke database and the PIM database to locate corresponding entries in each database that indicate that a particular one of the stroke entries was created at a time that corresponds to a particular one of the PIM database entries. The controller is responsive to the search being successful for forming a link between the stroke entry and the PIM database entry for enabling corresponding stroke information to be visualized in association with the particular one of the PIM database entries.

Abstract: A novel optoelectronic packaging submount arrangement which incorporates a 90° C. electrical conductor turn, and more specifically methods of producing optoelectronic packaging submount arrangement incorporating 90° C. electrical conductor turns.

Abstract: Apparatus for correcting speech including one or more words of a predetermined language comprises candidate word correlating means for correlating each of one or more speech data items of words to one or more candidate words obtained by recognizing said speech data items indicating the words. Analogous word correlating means correlates each of the candidate words correlated to the speech data items to null or more analogous words which may correspond to a pronunciation of each of the candidate words. The speech correcting apparatus further comprises pronunciation correcting data output means for outputting pronunciation correcting data corresponding to the analogous word indicated by the speech data item and correcting the pronunciation of the word indicated by the speech data item when the word indicated by the speech data item matches the analogous word correlated to each of the candidate words which are correlated to the speech data item.

Abstract: Methods and apparatus are provided for retrieving audio information based on the audio content as well as the identity of the speaker. The results of content and speaker-based audio information retrieval methods are combined to provide references to audio information (and indirectly to video). A query search system retrieves information responsive to a textual query containing a text string (one or more key words), and the identity of a given speaker. An indexing system transcribes and indexes the audio information to create time-stamped content index file(s) and speaker index file(s). An audio retrieval system uses the generated content and speaker indexes to perform query-document matching based on the audio content and the speaker identity. Documents satisfying the user-specified content and speaker constraints are identified by comparing the start and end times of the document segments in both the content and speaker domains.

Abstract: An audio retrieval system and method are provided for augmenting the transcription of an audio file with one or more alternate word or phrase choices, such as next-best guesses for each word or phrase, in addition to the best word sequence identified by the transcription process. The audio retrieval system can utilize a primary index file containing the best identified words and/or phrases for each portion of the input audio stream and a supplemental index file containing alternative choices for each word or phrase in the transcript. The present invention allows words that are incorrectly transcribed during speech recognition to be identified in response to a textual query by searching the supplemental index files. During an indexing process, the list of alternative word or phrase choices provided by the speech recognition system are collected to produce a set of supplemental index files.

Abstract: To provide an improved information processing apparatus and a method for controlling the same, which enables to smoothly transfer data, such as processed results obtained from execution of an application program, an HTML file acquired from a Web server in accordance with the TCP/IP protocol or the like, to an external device (PDA) by using an infrared communication function. The disclosed information processing apparatus periodically accesses a predetermined server machine (e.g., a Web server) to acquire a desired file (e.g., an HTML file). This file acquisition operation is carried out without the involvement of operations of an infrared transceiver. In other words, the information processing apparatus attempts to continually perform caching of the most recent download data.