Abstract: This invention provides a signal processing apparatus for improving the speech determination accuracy in an input sound. The signal processing apparatus includes a transformer that transforms an input signal into an amplitude component signal in a frequency domain, a calculator that calculates a norm of a change in the amplitude component signal in a frequency direction, an accumulator that accumulates the norm of the change in the amplitude component signal in the frequency direction calculated by the calculator, and an analyzer that analyzes speech in the input signal in accordance with an accumulated value of the norm of the change in the amplitude component signal in the frequency direction calculated by the accumulator.

Abstract: Provided are a device and a method for transmitting and receiving voice data in a wireless communication system. A method for operating a transmission terminal for transmitting a voice signal comprises the steps of: generating sampling and bitrate request information including sampling information for determining a sampling rate of the voice signal and bitrate information for determining a bitrate of the voice signal, and transmitting the generated sampling and bitrate request information to a reception terminal; receiving, from the reception terminal, combined determination information obtained by at least one combination of the sampling rate determined on the basis of the sampling information and the bitrate determined on the basis of the bitrate information; and compressing the voice signal according to the received combined determination information, and transmitting the compressed voice signal to the reception terminal.

Abstract: The methods, apparatus, and systems described herein are designed to reduce false positive voice print matching with fraudulent callers. A voice print of a call is created and compared to known voice prints to determine if it matches one or more of the known voice prints, and to transaction data associated with a database of voice prints. The methods include a pre-processing step to separate speech from non-speech, selecting a number of elements that affect the voice print the most, and/or generating a first score based on the number of selected audio elements matching audio elements of a voice print from the plurality of fraudulent speakers, determining if the first score exceeds a predetermined threshold score for the fraudulent speaker, and comparing the selected audio elements for the unknown caller, where the score exceeds the predetermined threshold score, to the voice prints associated with the customer account.

Abstract: The present invention provides a method that includes receiving a result word set in a target language representing a translation of a test word set in a source language. When the result word set is not in a set of acceptable translations, the method includes measuring a minimum number of edits to transform the result word set into a transform word set. The transform word set is in the set of acceptable translations. A system is provided that includes a receiver to receive a result word set and a counter to measure a minimum number of edits to transform the result word set into a transform word set. A method is provided that includes automatically determining a translation ability of a human translator based on a test result. The method also includes adjusting the translation ability of the human translator based on historical data of translations performed by the human translator.

Abstract: Techniques for controlling a voice activated feature of a voice activated device are described. Data from one or more sensors and data indicative of a status of a user are received. Based on the analyzing the data, a proximity of the user relative to the location of the voice activated device is determined. One or more voice activated features of the voice activated device are enabled based at least in part on the determined proximity, one or more rules, and one or more user preferences.

Abstract: A system and method for improving accuracy of natural language processing using a plurality of speech recognition engines, a data fusion model to identify a correct result from the plurality of speech recognition engines and a semantic understanding model, separate and distinct from the speech recognition model, to process the correct results. A corpus is developed using the correct results and the corpus is used to train the data fusion model and the semantic understanding model.

Abstract: The present invention relates to a system for a visually impaired user comprising a micro camera coupled to an eye wear, proximity sensors coupled to the wearable objects and the handheld electronic device coupled to both micro camera & proximity sensor and to the database. The visually impaired user instructs the micro camera and proximity sensors through his handheld electronic device to capture the data of proximally placed objects, places or people. The captured data of both the micro camera and the proximity sensors provides the complete information about the captured object, place or people and transmit it to the database of the handheld electronic device. This captured data is then processed by the database and the relevant audio output, corresponding to the captured data, is transmitted to the ear wear or the speaker of the handheld electronic device.

Abstract: The present disclosure describes systems and methods of fraud protection. A listening device may monitor ambient sound and detect one or more triggering sound patterns. The listening device may record the one or more triggering sound patterns and record a person's voice, and send the recordings to a server, wherein the server is configured to calculate a confidence level of fraud based on the one or more triggering sound patterns and the recording of the person's voice using a speech pattern model of the person, and to output an alert if the confidence level of fraud is greater than a threshold value.

Abstract: The present invention relates to a method and apparatus for improving spontaneous speech recognition performance. The present invention is directed to providing a method and apparatus for improving spontaneous speech recognition performance by extracting a phase feature as well as a magnitude feature of a voice signal transformed to the frequency domain, detecting a syllabic nucleus on the basis of a deep neural network using a multi-frame output, determining a speaking rate by dividing the number of syllabic nuclei by a voice section interval detected by a voice detector, calculating a length variation or an overlap factor according to the speaking rate, and performing cepstrum length normalization or time scale modification with a voice length appropriate for an acoustic model.

Abstract: A computer-implemented method can include receiving, at a computing device having one or more processors, a user input in a first writing system. The method can include determining one or more candidates based on the user input that correspond to the user input represented in a second writing system different than the first writing system. The method can include determining whether at least one of the candidates corresponds to a proper name of a geographic location and, when at least one of the candidates corresponds to the proper name of the geographic location, determining a local name of the geographic location. The local name can be the proper name of the geographic location in a local language associated with the geographic location. The method can include augmenting the one or more candidates to include the local name to obtain an augmented candidate list, and outputting the augmented candidate list.

Abstract: Various embodiments of systems, computer program products, and methods to assess translation quality are described herein. In an aspect, a translated text is received during translation of content from a source language to a target language. The received translated text is detected as an incorrect translation by analyzing a number of quality checks in a translation quality algorithm. An error category corresponding to the incorrect translation is determined based on a root cause of the incorrect translation. Further, a counter is incremented in a corresponding error category. When the counter exceeds a threshold, an action to prevent submission of the translation is triggered based on a combined error score. The combined error score is a combined weighted error score from error categories which have exceeded their respective thresholds.

Abstract: Examples of techniques text-to-speech pre-processing for speech recognition and speech synthesis are disclosed. In one example implementation, a computer-implemented method includes receiving, by a processing device, an automated speech recognition output comprising an n-best list and associated confidence scores. The method further includes performing, by the processing device, a TTS pre-processing on the n-best list and associated confidence scores to generate a read back message, wherein the read back message comprises a read back instruction. The method further includes sending, by the processing device, the read back message to a TTS speech synthesizer for generating an audible signal based on the read back message to cause an audio device to present the read back message.

Abstract: The present disclosure provides a method and a device for processing a speech based on artificial intelligence. The method includes: receiving a speech processing request, in which the speech processing request includes a first digital speech signal and a first sample frequency corresponding to the first digital speech signal; selecting a target speech processing model from a pre-trained speech processing model base according to the first sample frequency; performing up-sampling processing on the first digital speech signal using the target speech processing model to generate a second digital speech signal having a second sample frequency, in which the second sample frequency is larger than the first sample frequency.

Abstract: A method includes associating, for each one of a plurality of answer definitions, at least one or more pattern-form questions, wherein each answer definition has an associated jump target that defines a respective entry point into the workspace analytics system to provide information responsive to the associated one or more pattern-form questions. The method further includes receiving a user input including capturing input text defining a natural language user query, matching the received input text to one of the pattern-form questions thereby selecting the jump target associated with the matched pattern-form question, and generating a response to the natural language user query by retrieving information from the workspace analytics system by referencing a link based on the selected jump target and zero or more parameters values.

Abstract: A method for providing a translation, including: acquiring a first sentence in a first language via a user terminal; determining whether the first sentence is in a database including sentences in the first language and corresponding translations in a second language; if the first sentence is not in the database, generating second sentences by replacing one or more words in the first sentence, based on a predetermined rule; calculating respective degrees of coincidence for syntax between the second sentences and sentences in the first language included in the database; extracting third sentences in the first language which are included in the database and for which the calculated degree of coincidence is at least a threshold value; and displaying fourth sentences in the second language which are corresponding translations for the third sentences in the database, on the user terminal as corresponding translation references for the first sentence.

Abstract: A computer implemented method to generate a pattern matching machine to identify matches of a plurality of symbol patterns in a sequence of input symbols, wherein one or more of the symbol patterns includes a plurality of wildcard symbols.

Abstract: In one example, a method includes method comprising: receiving audio data generated by a microphone of a current computing device; identifying, based on the audio data, one or more computing devices that each emitted a respective audio signal in response to speech reception being activated at the current computing device; and selecting either the current computing device or a particular computing device from the identified one or more computing devices to satisfy a spoken utterance determined based on the audio data.

Abstract: Embodiments of the present invention address deficiencies of the art in respect to podcasting and provide a method, system and computer program product for speed podcasting. In an embodiment of the invention, a speed podcasting method can include speech recognizing an audio portion of a podcast, parsing the speech recognized audio portion to identify essential words, and playing back only audio segments and corresponding video segments of the podcast including the essential words while excluding from playback audio segments and corresponding video segments of the podcast including non-essential words.

Abstract: In a method of creating similar sentences from an entered original, one or more second phrases having the same meaning as a first phrase, which is part of the original, are extracted from a first database; an N-gram value is calculated according to a context dependence value, in a second database, corresponding to the one or more second phrases; one or more contiguous third phrases that include a number of second phrases equivalent to the N-gram value are extracted from one or more sentences obtained by replacing, in the original, the first phrase with the one or more second phrases; the appearance frequency of the one or more third phrases in a third database is calculated; and if the calculated appearance frequency is determined to be larger than or equal to a threshold, the one or more sentences are used as similar sentences of the original and are externally output.

Abstract: The present invention provides a method that includes receiving a result word set in a target language representing a translation of a test word set in a source language. When the result word set is not in a set of acceptable translations, the method includes measuring a minimum number of edits to transform the result word set into a transform word set. The transform word set is in the set of acceptable translations. A system is provided that includes a receiver to receive a result word set and a counter to measure a minimum number of edits to transform the result word set into a transform word set. A method is provided that includes automatically determining a translation ability of a human translator based on a test result. The method also includes adjusting the translation ability of the human translator based on historical data of translations performed by the human translator.