Abstract: Systems and methods of validating transcriptions of natural language content using crowdsourced validation jobs are provided herein. In various implementations, a transcription pair comprising natural language content and text corresponding to a transcription of the natural language content may be gathered. A group of validation devices may be selected for reviewing the transcription pair. A crowdsourced validation job may be created for the group of validation devices. The crowdsourced validation job may be provided to the group of validation devices. One or more votes representing whether or not the text accurately represents the natural language content may be received from the group of validation devices. Based on the one or more votes received, the transcription pair may be stored in a validated transcription library, which may be used to process end-user voice data.

Abstract: A Transcription Engine is configured to broadcast over the Internet and stream text associated with the broadcast to registered and authenticated end users who may be hearing impaired or may have difficulty understanding the language used in the broadcast. The Transcription Engine comprises an authentication server, a database server and a Transcription server. End users are authenticated automatically by the Transcription Engine prior to being given access to a live or recorded broadcast.

Abstract: Systems and methods for checking the consistency of established facts within internal works according to the present disclosure operate by identifying established facts within the internal works and determining whether any of the established facts are contradictory to one another. Facts may be established and conflicts may be identified by any means, such as by determining associations between words of the internal work, or by consulting one or more external resources. If a contradiction between established facts is identified, then an author of the internal work or other user may be notified, and a change to the internal work may be recommended to the author or user, or requested from the author or user.

Abstract: Systems, methods, and devices for intelligent speech recognition and processing. According to one embodiment, a method for improving intelligibility of a speech signal may include (1) at least one processor receiving an incoming speech signal comprising a plurality of sound elements; (2) the at least one processor recognizing a sound element in the incoming speech signal to improve the intelligibility thereof; (3) the at least one processor processing the sound element by at least one of modifying and replacing the sound element; and (4) the at least one processor outputting the processed speech signal comprising the processed sound element.

Abstract: A method and an electronic device for handling a spoken user request of a user, executable by each one of at least two applications installed on the electronic device, comprising determining that the spoken user request corresponds to an action executable by an other one of the at least two of the applications; and causing execution of the action by the other one of the at least two of the applications. A method and an electronic device for handling a spoken user request received from a user of the electronic device, comprising detecting reception of a spoken user request by a first application; transferring the spoken user request to the second application by the first application; determining, by the second application, that the spoken user request corresponds to an action executable by a third application; and causing, by the second application, execution of the action by the third application.

Abstract: A base station for use in a communication system is configured to determine whether a set of access information provided by an end unit is acceptable to the base station. The access information facilitates identification by the base station of one or more codecs used by the end unit to encode and decode digital audio signals. Other end units may use different codecs to encode and decode digital audio signals. The base station is configured to simultaneously handle multiple different codecs. The communication system also comprises at least one end unit configured to determine the codec used by the base station, request registration to the base station, and set the determined codec as the codec to use for communications with the base station.

Abstract: Systems, methods, and devices for intelligent speech recognition and processing are disclosed. According to one embodiment, a method for improving intelligibility of a speech signal may include (1) at least one processor receiving an incoming speech signal comprising a plurality of sound elements; (2) the at least one processor recognizing a sound element in the incoming speech signal to improve the intelligibility thereof; (3) the at least one processor processing the sound element by at least one of modifying and replacing the sound element; and (4) the at least one processor outputting the processed speech signal comprising the processed sound element.

Abstract: Systems for improving or generating a spoken language understanding system using a multitask learning method for intent or call-type classification. The multitask learning method aims at training tasks in parallel while using a shared representation. A computing device automatically re-uses the existing labeled data from various applications, which are similar but may have different call-types, intents or intent distributions to improve the performance. An automated intent mapping algorithm operates across applications. In one aspect, active learning is employed to selectively sample the data to be re-used.

Abstract: A method for recognizing an audio sample locates an audio file that matches the sample from a database indexing original recordings. Each indexed file is represented in the database index by a set of landmark timepoints and associated fingerprints. Landmarks occur at reproducible locations within the file, while fingerprints represent features of the signal at or near the landmark timepoints, and the fingerprints are invariant with respect to time stretching of the sample. To perform recognition, landmarks and fingerprints are computed for the unknown sample and used to retrieve matching fingerprints from the database. For each file containing matching fingerprints, the landmarks are compared with landmarks of the sample at which the same fingerprints were computed. If a large number of corresponding landmarks are linearly related, i.e., if equivalent fingerprints of the sample and retrieved file have the same time evolution, then the file is identified with the sample.

Abstract: A method of performing gain adjustment in an electronic device includes determining a first set of spectral frequency values and determining a second set of spectral frequency values. The first set of spectral frequency values corresponds to a high-band portion of an audio signal received at the electronic device. The second set of spectral frequency values approximates the first set of spectral frequency values in the high band portion of the audio signal. The method includes estimating a spectral distortion corresponding to a difference between the first set of spectral frequency values and the second set of spectral frequency values and adjusting, based on the spectral distortion, a gain value corresponding to at least a portion of the audio signal. The method also includes transmitting an encoded bitstream that includes information corresponding to the adjusted gain value and the second set of spectral frequency values.

Abstract: The present disclosure relates to systems, methods, and computer-readable media for generating a lexicon for use with speech recognition. The method includes overgenerating potential pronunciations by converting portions of symbolic input into a number of possible lexical pronunciation variants based on an established set of conversion rules, wherein the symbolic input comprises labeled speech data and selecting pronunciations in a speech recognition context from the potential pronunciations, to yield selected potential pronunciations. The method further includes retraining the established set of conversion rules based on the selected potential pronunciations.

Abstract: A speech recognition method, a speech recognition device, and an electronic device. In this method, first determining is performed by using a sample environment corresponding to a detection speech and a previous environment type, so as to output a corresponding speech correction instruction to a speech engine; then, a to-be-recognized speech is input to the speech engine and a noise type detection engine at the same time, and the speech engine corrects the to-be-recognized speech by using the speech correction instruction, so that quality of an original speech is not impaired by noise processing, and a corresponding initial recognition result is output; the noise type detection engine determines a current environment type by using the to-be-recognized speech and a speech training sample under a different environment; finally, confidence of the initial recognition result is adjusted by using the current environment type.

Abstract: Customer contact handling at a customer service system. Contact text and contact metadata is received from a user communication device. Semantic characteristics of the contact text are determined based on semantic analysis of the contact text. A user profile is identified based on the contact metadata. User data associated with the identified user profile is retrieved. A contact need classification is determined based on the semantic characteristics and user data. A service agent profile from a plurality of service agent profiles is selected based on one or more of the contact need classification, semantic characteristics, and the user data. Contact event data is provided to an agent communication device associated with the selected agent profile.

Abstract: Systems, methods, and products for language learning that may determine a level score of a learner based upon responsive inputs received from the learner during sessions. A pedagogical value threshold is determined based upon the level score of the learner and a resource difficulty score of a resource containing text having one or more words. Each word having a word difficulty score satisfying the pedagogical value threshold is stored into a non-transitory machine-readable storage media keyword store.

Abstract: A display apparatus includes an input unit configured to receive a user command; an output unit configured to output a registration suitability determination result for the user command; and a processor configured to generate phonetic symbols for the user command, analyze the generated phonetic symbols to determine registration suitability for the user command, and control the output unit to output the registration suitability determination result for the user command. Therefore, the display apparatus may register a user command which is resistant to misrecognition and guarantees high recognition rate among user commands defined by a user.

Abstract: Embodiments of the present application include an apparatus with at least one processor that is configured to determine whether to electronically publish textual data in an online environment (e.g., the Internet) based on a comparison of a computed numeric likelihood to a set of threshold values. The set of threshold values may include a first threshold value representative of numeric likelihoods assigned to a first portion of reference content suitable for publication in the online environment, and a second threshold value representative of numeric likelihoods assigned to a second portion of the reference content unsuitable for publication in the online environment. The second threshold value may exceed the first threshold value.

Abstract: [Object] An object is to provide an easy-to-use speech processing system attaining higher accuracy of speech recognition. [Solution] Receiving a speech utterance, the speech processing system performs speech recognition and displays a text of the recognition result. Further, the speech processing system translates the recognition result in accordance with settings to a text of another language and displays and synthesizes speech of the translated result. Further, the speech processing system selects utterance candidates having high possibility to be uttered as the next utterance and having high translation and speech recognitions scores, using outputs of various sensors at the time of utterance, a pre-trained utterance sequence model and translation and speech recognition scores of utterance candidates, and recommends utterance candidates in the form of an utterance candidate recommendation list. A user can think of what to say next using the utterances in utterance candidate recommendation list as hints.

Abstract: An encoder for encoding an audio signal includes an analyzer for analyzing the audio signal and for determining analysis prediction coefficients from the audio signal. The encoder includes a converter for deriving converted prediction coefficients from the analysis prediction coefficients, a memory for storing a multitude of correction values and a calculator. The calculator includes a processor for processing the converted prediction coefficients to obtain spectral weighting factors. The calculator includes a combiner for combining the spectral weighting factors and the multitude of correction values to obtain corrected weighting factors. A quantizer of the calculator is configured for quantizing the converted prediction coefficients using the corrected weighting factors to obtain a quantized representation of the converted prediction coefficients.

Abstract: Systems, methods and apparatus for generating, distributing, and using speech recognition models. A request including a speech data representation of an utterance or feature data extracted from the speech data representation of the utterance is received at a computing device. A transcription of the utterance by applying a speech recognition model to the speech data representation of the utterance or the feature data extracted from the speech data representation of the utterance is obtained by the computing device. A keyword that occurs in the transcription of the utterance is identified by the computing device. Communication between communication devices based at least on identifying the keyword is initiated initiating by the computing device.

Abstract: An audio decoder device for decoding a bitstream includes a bitstream receiver configured to receive the bitstream and to derive an encoded audio signal from the bitstream; a core decoder module configured for deriving a decoded audio signal in a time domain from the encoded audio signal; a temporal envelope generator configured to determine a temporal envelope of the decoded audio signal; a bandwidth extension module configured to produce a frequency domain bandwidth extension signal; a time-to-frequency converter configured to transform the decoded audio signal into a frequency domain decoded audio signal; a combiner configured to combine the frequency domain decoded audio signal and the frequency domain bandwidth extension signal in order to produce a bandwidth extended frequency domain audio signal; and a frequency-to-time converter configured to transform the bandwidth extended frequency domain audio signal into a bandwidth-extended time domain audio signal.