Abstract: An interactive speech interface unit includes speech recognizer for recognizing input speech of user utterance and converting the recognized input speech into a character string; an input statement analyzer means for analyzing the character string and converting the analyzed character string into semantic representation; an interactive controller for controlling flow of an interactive status and accessing an application; an output statement generator for generating an intermediate language to be output to the user; a speech generator for converting the intermediate language into speech and outputting the speech; and an application interface for accessing the application using the semantic representation output from the interactive controller, wherein the interactive controller puts series of interactive sequences having calling relations together in a plurality of interactive tasks in association with relations and includes an interactive task hierarchical data base for storing the interactive tasks in a hier

Abstract: A speech recognition system plays prompts to a user in order to obtain information from the user. If the user begins to speak, the prompt should stop. However, the system may receive sounds other than speech from the user while playing a prompt, in which case the prompt should continue. The system temporarily stops a prompt when it detects a sound or when it preliminarily determines that a detected sound may be a target sound (such as words from the user). The system then determines whether the received sound is a target sound or some other sound (such as coughing or a door shutting). If the received sound is not determined to be a target sound, then the prompt is resumed. The prompt can be resumed at any appropriate point, such as the point where it was stopped, a prior phrase boundary, or the beginning of the prompt.

Abstract: Computer method and apparatus for extracting information from a Web page is disclosed. The invention apparatus is formed of an extractor coupled to receive Web pages from a source. The extractor uses natural language processing to extract desired information from the Web page. A storage subsystem receives from the extractor the extracted desired information and stores the extracted desired information in a database. The invention method for extracting data from a Web page includes the computer implemented steps of (i) using natural language processing, finding possible formal names on a given Web page, (ii) using pattern matching, searching the given Web page for formal names not found by the natural language processing, and (iii) refining a combined set of the found formal names to produce a working set of people and organization names extracted from the given Web page. The refining includes determining aliases of respective people and organization names, so as to effectively reduce duplicate names.

Abstract: The invention comprises capturing a customer's speech, recognizing a key word in the customer's speech, searching a database, and retrieving information from the database. The retrieving is a real-time process, completed during a conversation involving the customer and a customer service representative. Examples include methods employing computerized speech recognition and key words to improve customer service, systems for executing methods of the present invention, and instructions on a computer-usable medium, or resident in a computer system, for executing methods of the present invention.

Abstract: The present invention relates to a foreign language training apparatus and the method. The purpose of the invention is to provide a foreign language training apparatus and the method that can enhance the efficiency of studying foreign languages. The invention provides the apparatus comprising: storage including a plurality of multimedia files for learning foreign languages and a program for executing the multimedia files; a checking means for checking executing time of the multimedia files; an input for inputting a control signal; a controller that selects a first file of the group of files from the storage, executes the first file using the program, selects and executes a second file after a predetermined time dependent on the executing time of the first file checked by the checking means according to the control signal; and an output for outputting the executed multimedia files.

Abstract: A system and method of providing sender customization of multi-media messages through the use of inserted images or video. The images or video may be sender-created or predefined and available to the sender via a web server. The method relates to customizing a multi-media message created by a sender for a recipient, the multi-media message having an animated entity audibly presenting speech converted from text created by the sender. The method comprises receiving at least one image from the sender, associating each at least one image with a tag, presenting the sender with options to insert the tag associated with one of the at least one image into the sender text, and after the sender inserts the tag associated with one of the at least one images into the sender text, delivering the multi-media message with the at least one image presented as background to the animated entity according to a position of the tag associated with the at least one image in the sender text.

Abstract: A Sound on Sound-Annotations (SOS-A) system facilitates the collection, categorization, and retrieval of streams of sound. A stream of sound is captured and annotations of sound concerning the stream of sound are generated for positions of interest or relevancy. The annotations add additional information concerning the stream of sound at the points of interest. Markers of sound are logically or physically inserted in the stream of sound to identify the locations associated with the annotations of sound. The markers of sound point to or link the annotation of sound. The annotations of sound are also captured and can convey any information desired; for example, add description, provide evidence, challenge the validity, ask questions, etc. Any form or frequency of sound can be utilized with the stream of sound, the marker of sound, and/or the annotations of sound.

Abstract: A method and computer-readable medium are provided that construct a collocation mistake pattern database for use in writing in a first language by a person whose native language is a second language. The method includes obtaining a bilingual corpus having sentences in first and second languages and extracting second language word pairs from the second language sentences in the corpus. For each second language word pair extracted from the corpus, a corresponding first language word pair is extracted from the corresponding first language sentence in the corpus to determine a correct first language translation for the second language word pair. Also, for each second language word pair extracted from the corpus, a set of combinations of first language translation words corresponding to the second language word pair is created.

Abstract: An improved representation of transients in audio signals comprises modifying transient locations in such a way that a transient can occur only at a beginning of a sinusoidal segment. The modification procedure comprises the steps: detecting a beginning and an end of a transient using an energy-based approach with two sliding rectangular windows; moving samples between the beginning and the end of the transient to the locations specified by the segmentation used; and time-warping the signal parts in between the transients in order to fill the intervals between the modified transients.

Abstract: A speech synthesis apparatus (10) comprises speech segment disassembling means (101) for disassembling the speech segments each including at least one phoneme into a plurality of pitch waveforms, phase characteristic transforming means (103) for transforming the phase characteristics of the pitch waveforms into a uniformed phase characteristic, pitch waveform classifying means (104) for classifying the pitch waveforms into a plurality of groups, pitch waveform registering means (106) for registering the pitch waveforms in the database (111) by extracting one pitch waveform from among the pitch waveforms in each of the groups, and synthesizing means (107) for synthesizing the speech with the pitch waveforms registered in the database (111). The speech synthesis apparatus (10) thus constructed can synthesize a natural speech using a relatively small database capacity.

Abstract: A digital audio signal to be replayed is processed in a waveform thereof. A frequency bandwidth of the audio signal is expanded through conversion of a sampling frequency, and then the audio signal is low-pass-filtered with a low-pass cut-off frequency corresponding to the converted sampling frequency. An interval of time between two waveform peaks of the audio signal is detected, and then difference data between current data of the audio signal and past data thereof is calculated. The difference data are subject to weighting depending on the interval, and then output data are produced based on both the low-pass-filtered audio signal and the weighted difference data. This processing, which can be realized by activation of software, improves audio quality when compressed audio data is replayed.

Abstract: A method, system, and program for origin device initiated caller identification are provided. In response to detecting a call extended to a destination device, extending a request from said destination device to an origin device requesting a voice utterance of the caller at said origin device. A caller identity associated with the voice utterance is identified at the destination device, such that a callee receiving the call at the destination device is informed of the caller identity before choosing whether to speak with the caller.

Abstract: A probability distribution for speech model parameters, such as auto-regression parameters, is used to identify a distribution of denoised values from a noisy signal. Under one embodiment, the probability distributions of the speech model parameters and the denoised values are adjusted to improve a variational inference so that the variational inference better approximates the joint probability of the speech model parameters and the denoised values given a noisy signal. In some embodiments, this improvement is performed during an expectation step in an expectation-maximization algorithm. The statistical model can also be used to identify an average spectrum for the clean signal and this average spectrum may be provided to a speech recognizer instead of the estimate of the clean signal.

Abstract: A method, system, computer program product, and method of doing business by providing improved audio compression wherein an audio stream is securely transformed to an encoded text stream (such as an ASCII, EBCDIC, or Unicode text stream). One or more components which are involved in the transformation process are authenticated. A unique identifier of each such component is included within cryptographically-protected information that is provided for the encoded text stream. A digital signature is preferably used for the cryptographic protection, thereby digitally notarizing the encoded text stream. The authenticity and integrity of the encoded text stream can therefore be verified. In preferred embodiments, the authenticated identities of components performing the transformation can also be determined from the cryptographically-protected information.

Abstract: A system and method of providing sender-customization of multi-media messages through the use of emoticons is disclosed. The sender inserts the emoticons into a text message. As an animated face audibly delivers the text, emoticons associated with the message are started a predetermined period of time or number of words prior to the position of the emoticon in the message text and completed a predetermined length of time or number of words following the location of the emoticon. The sender may insert emoticons through the use of emoticon buttons that are icons available for choosing. Upon sender selections of an emoticon, an icon representing the emoticon is inserted into the text at the position of the cursor. Once an emoticon is chosen, the sender may also choose the amplitude for the emoticon and increased or decreased amplitude will be displayed in the icon inserted into the message text.

Abstract: A Hidden Markov model is used to segment a data sequence. To reduce the potential for error that may result from the Markov assumption, the Viterbi dynamic programming algorithm is modified to apply a multiplicative factor if a particular set of states is re-entered. As a result, structural domain knowledge is incorporated into the algorithm by expanding the state space in the dynamic programming recurrence. In a specific example of segmenting resumes, the factor is used to reward or penalize (even require or prohibit) a segmentation of the resume that results in the re-entry into a section such as Experience or Contact Information. The method may be used to impose global constraints in the processing of an input sequence or to impose constraints to local sub-sequences.

Abstract: A speech recognition interface and computer-readable medium build a grammar for speech recognition that reduces latency in speech-based applications. The interface and medium receive instructions to add a new phrase and semantic information to a grammar. The new phrase is combined with at least one other phrase in the grammar to form a composite grammar structure. The semantic information is then associated with a single word or transition in the grammar structure by selecting the first possible word or transition that semantically differentiates the new phrase from all other phrases in the grammar structure. By placing the semantic information in this position, the semantic information is placed as far forward in the grammar as possible without introducing semantic ambiguity into the grammar structure.

Abstract: A scalable audio codec processes, quantizes and encodes audio signals into an embedded audio bitstream of bit-planes each having a data unit. The data unit has a beginning refinement bits partition, a second significance bits partition, a third sign boundary mark bits partition, and a fourth sign bits partition. The second and fourth partitions form a boundary for the third partition. The quantizing uses a variable length coding algorithm. The third partition is an invalid codeword for a predetermined encoding method being used to encode. The codec uses a decoder to decode the embedded audio bitstream of bit-planes using Reversible exponential Golomb (Exp-Golomb) codes in a Reversible Variable Length Code (RVLC) algorithm to produce quantized data of weighted subbands. An inverse quantizer dequantizes the quantized data into audio signals.