Abstract: Speech recognition systems may perform the following operations: receiving audio; recognizing the audio using language models for different languages to produce recognition candidates for the audio, where the recognition candidates are associated with corresponding recognition scores; identifying a candidate language for the audio; selecting a recognition candidate based on the recognition scores and the candidate language; and outputting data corresponding to the selected recognition candidate as a recognized version of the audio.

Abstract: Methods and apparatuses for wearing state device operation are disclosed. In one example, a headset includes a sensor for detecting a headset donned state or a headset doffed state. The headset operation is modified based on whether the headset is donned or doffed.

Abstract: A procurement system may include a first interface configured to receive a query from a user, a command module configured to parameterize the query, an intelligent search and match engine configured to compare the parameterized query with stored queries in a historical knowledge base and, in the event the parameterized query does not match a stored query within the historical knowledge base, search for a match in a plurality of knowledge models, and a response solution engine configured to receive a system response ID from the intelligent search and match engine, the response solution engine being configured to initiate a system action by interacting with sub-system and related databases to generate a system response.

Abstract: According to one embodiment, a speech synthesis apparatus includes a language analyzer, statistical model storage, model selector, parameter generator, basis model storage, and filter processor. The language analyzer analyzes text data and outputs language information data that represents linguistic information of the text data. The statistical model storage stores statistical models prepared by statistically modeling acoustic information included in speech. The model selector selects a statistical model from the models based on the language information data. The parameter generator generates speech parameter sequences using the statistical model selected by the model selector. The basis model storage stores a basis model including basis vectors, each of which expresses speech information for each limited frequency range. The filter processor outputs synthetic speech by executing filter processing of the speech parameter sequences and the basis model.

Abstract: One-to-many comparisons of callers' voice prints with known voice prints to identify any matches between them. When a customer communicates with a particular entity, such as a customer service center, the system makes a recording of the real-time call including both the customer's and agent's voices. The system segments the recording to extract at least a portion of the customer's voice to create a customer voice print, and it formats the segmented voice print for network transmission to a server. The server compares the customer's voice print with multiple known voice prints to determine any matches, meaning that the customer's voice print and one of the known voice prints are likely from the same person. The identification of any matches can be used for a variety of purposes, such as determining whether to authorize a transaction requested by the customer.

Abstract: Techniques for determining one or more preferred languages for a user are provided. The preferred languages may be determined based upon a set of language indicators. The language indicators are analyzed using, for example, rules-based techniques, clustering, language classifiers, and the like, or combinations thereof. Language indicators can include or be derived from information about the user's behavior, location, preferences, social connections, or other data related to the user.

Abstract: Embodiments of the present invention provide an audio signal coding and decoding method and device. The coding method includes: dividing a frequency band of an audio signal into a plurality of sub-bands, and quantifying a sub-band normalization factor of each sub-band; determining signal bandwidth of bit allocation according to the quantized sub-band normalization factor, or according to the quantized sub-band normalization factor and bit rate information; allocating bits for a sub-band within the determined signal bandwidth; and coding a spectrum coefficient of the audio signal according to the bits allocated for each sub-band. According to embodiments of the present invention, during coding and decoding, signal bandwidth of bit allocation is determined according to the quantized sub-band normalization factor and bit rate information. In this manner, the determined signal bandwidth is effectively coded and decoded by centralizing the bits, and audio quality is improved.

Abstract: A speech recognition apparatus includes a first recognition dictionary, a speech input unit, a speech recognition unit, a speech transmission unit, a recognition result receipt unit, and a control unit. The speech recognition unit recognizes a speech based on a first recognition dictionary, and outputs a first recognition result. A server recognizes the speech based on a second recognition dictionary, and outputs a second recognition result. The control unit determines a likelihood level of a selected candidate obtained based on the first recognition result, and accordingly controls an output unit to output at least one of the first recognition result and the second recognition result. When the likelihood level of the selected candidate is equal to or higher than a threshold level, the control unit controls the output unit to output the first recognition result irrespective of whether the second recognition result is received from the server.

Abstract: An Internet radio station program discovery service is provided. A plurality of Internet radio station programs is obtained. For each Internet radio station program of the plurality of Internet radio station programs, the Internet radio station program is dynamically categorized by mapping a dynamically identified topic of the Internet radio station program to a content classifier. A User is enabled to discover an Internet radio station program of interest from the plurality of Internet radio station programs based on the dynamic categorizations for the plurality of Internet radio station programs.

Abstract: An automatic speech recognition engine may generate text or tokens that correspond to audio data. For example, the automatic speech recognition engine may generate first text or first speech tokens corresponding to a first portion of audio data. The automatic speech recognition engine may further generate second text or second speech tokens that correspond to a first portion of the audio data and a second portion of the audio data. The text or speech tokens generated by the automatic speech recognition engine may be provided to a device for presentation thereon. In some embodiments, the automatic speech recognition engine generates the second text or second speech tokens substantially while the first text or first speech tokens are presented on the device.

Abstract: A method is disclosed herein for recognizing a repeated utterance in a mobile computing device via a processor. A first utterance is detected being spoken into a first mobile computing device. Likewise, a second utterance is detected being spoken into a second mobile computing device within a predetermined time period. The second utterance substantially matches the first spoken utterance and the first and second mobile computing devices are communicatively coupled to each other. The processor enables capturing, at least temporarily, a matching utterance for performing a subsequent processing function. The performed subsequent processing function is based on a type of captured utterance.

Abstract: The subject matter of this specification can be implemented in a computer-implemented method that includes receiving utterances and transcripts thereof. The method includes analyzing the utterances and transcripts to determine certain attributes, such as distances between prosodic contours for pairs of utterances. A model can be generated that can be used to estimate a distance between a determined prosodic contour for a received utterance and an unknown prosodic contour for a synthesized utterance when given a distance between attributes for text associated with the received utterance and the synthesized utterance.

Abstract: For generating privacy, a detection module detects an optical lingual cue from user speech that comprises an audible signal. A generation module transmits an inverse audible signal generated from the optical lingual cue.

Abstract: Regarding audio data related to document data, an image processing apparatus pertaining to the present invention generates text data by using a speech recognition technology in advance, and determines delimiter positions in the text data and the audio data in correspondence. In a keyword search, if a keyword is detected in the text data, the image processing apparatus plays the audio data from a delimiter that is immediately before the keyword.

Abstract: Methods, apparatus, and products are disclosed for adjusting a speech engine for a mobile computing device based on background noise, the mobile computing device operatively coupled to a microphone, that include: sampling, through the microphone, background noise for a plurality of operating environments in which the mobile computing device operates; generating, for each operating environment, a noise model in dependence upon the sampled background noise for that operating environment; and configuring the speech engine for the mobile computing device with the noise model for the operating environment in which the mobile computing device currently operates.

Abstract: A perplexity calculation device 500 includes: a weight coefficient calculating part 501 for, with respect to each of a plurality of text constituent words constituting a text, calculating a weight coefficient for correcting a degree of ease of word appearance having a value which becomes larger as a probability of appearance of the text constituent word becomes higher based on a statistical language model showing probabilities of appearance of words, based on word importance representing a degree of importance of the text constituent word; and a perplexity calculating part 502 for calculating perplexity of the statistical language model to the text, based on the calculated weight coefficients and the degrees of ease of word appearance.

Abstract: A method for using speech analysis to detect speech pathologies can begin with registration of a patient with a speech-based health monitor. A speech segment baseline representing an initial state of the patient's speech system can be established for the patient. When prompted, the patient can submit a speech segment representing a current state of the patient's speech system to the speech-based health monitor. The speech-based health monitor can analyze the submitted speech segment using the established speech segment baseline and/or a speech segment history that comprises speech segments previously submitted by the patient. Based upon said analysis, satisfaction of a health alert definition can be determined. A health alert definition can define an action performed by the speech-based health monitor when its associated triggering conditions are satisfied. The action associated with the at least one satisfied health alert definition can then be executed.

Abstract: The present disclosure provides an audio quantization coding and decoding device and a method thereof. In the method, before a quantization coding process is performed on a digital signal, the signal is pre-processed, the digital signal is split into multiple frames based on positive and negative half periods of the signal, and all audio data between two adjacent zero-crossing points belongs to the same positive and negative half periods, so as to have the same sign-bit. A pre-processing module groups the numeric data belonging to the same positive and negative half periods into the same frame. When coding, an audio quantization coding module only needs to record a sign-bit of the frame at a head of the frame, so the sign-bit of each batch of voice data in the frame may be omitted to reduce a data amount or improve a resolution of each batch of voice data.

Abstract: Clients connecting to a VoiceXML browser obtain a control channel. Using this channel, clients may initialize a new VoiceXML session or attach to an existing VoiceXML session. The client after obtaining a session may perform a range of actions including controlling and monitoring actions.

Abstract: An apparatus for processing an audio signal to obtain control information for a speech enhancement filter has a feature extractor for extracting at least one feature per frequency band of a plurality of frequency bands of a short-time spectral representation of a plurality of short-time spectral representations, where the at least one feature represents a spectral shape of the short-time spectral representation in the frequency band. The apparatus additionally has a feature combiner for combining the at least one feature for each frequency band using combination parameters to obtain the control information for the speech enhancement filter for a time portion of the audio signal. The feature combiner can use a neural network regression method, which is based on combination parameters determined in a training phase for the neural network.