Abstract: The present invention provides for improving training dataset by identifying errors in training dataset and generating improvement recommendations. In operation, the present invention provides for identifying and correcting duplicate utterances in training dataset comprising utterances-intent pairs. Further, a plurality of Natural Language ML models are trained with the corrected training dataset to obtain diverse set of trained ML models. Each utterance of training dataset are fed as input to trained ML models, and a probability of error associated with each utterances-intent pairs of training dataset are evaluated based on analysis of respective intent predictions received from each of the trained ML models. Furthermore, spelling errors in the dataset are identified and data-imbalances in the training dataset are evaluated.

Abstract: A multi-channel signal encoding method includes determining a downmixed signal of a first channel signal and a second channel signal, determining an initial reverberation gain parameter of the first channel signal and the second channel signal, determining a target reverberation gain parameter of the first channel signal and the second channel signal based on a correlation between the first channel signal and the downmixed signal, a correlation between the second channel signal and the downmixed signal, and the initial reverberation gain parameter, quantizing the first channel signal and the second channel signal based on the downmixed signal and the target reverberation gain parameter, and writing a quantized first channel signal and a quantized second channel signal into a bitstream.

Abstract: The present disclosure generally relates to a system and method for obtaining a diagnosis of a mental health condition. An exemplary system can receive an audio input; convert the audio input into a text string; identify a speaker associated with the text string; based on at least a portion of the audio input, determine a predefined audio characteristic of a plurality of predefined audio characteristics; based on the determined audio characteristic, identify an emotion corresponding to the portion of the audio input; generate a set of structured data based on the text string, the speaker, the predefined audio characteristic, and the identified emotion; and provide an output for obtaining the diagnosis of the mental disorder or condition, wherein the output is indicative of at least a portion of the set of structured data.

Abstract: A method manages bot dialogue. A user input is converted to a phrase vector. A set of identified tokens are identified by a token identification engine from the phrase vector. An unsupervised token is selected from the set of identified tokens. A supervised token is selected from the set of identified tokens. A voted token selected from the unsupervised token and the supervised token. A next token is identified based on a set of recent tokens that includes the voted token. The next token is presented as one of a voice communication and an email communication.

Abstract: A media edit point selection process can include a media editing software application programmatically converting speech to text and storing a timestamp-to-text map. The map correlates text corresponding to speech extracted from an audio track for the media clip to timestamps for the media clip. The timestamps correspond to words and some gaps in the speech from the audio track. The probability of identified gaps corresponding to a grammatical pause by the speaker is determined using the timestamp-to-text map and a semantic model. Potential edit points corresponding to grammatical pauses in the speech are stored for display or for additional use by the media editing software application. Text can optionally be displayed to a user during media editing.

Abstract: A method for extracting information from a document, comprising: receiving an identification of an entity to be extracted from the document; identifying candidates from the document, each candidate corresponding to a given element contained in the document and having a given location within the document; embedding the candidates, thereby obtaining an embedding vector for each candidate; for each candidate, comparing in a semantic space the respective embedding vector to previous embedding vectors associated with previous entity values previously chosen for the entity, thereby obtaining a first comparison result; for each candidate, comparing in a pixel space the given location within the document of the candidate to a location associated with the previous entity values previously chosen for the entity, thereby obtaining a second comparison result; sorting the candidates using the first and second comparison results obtained for each candidate, thereby obtaining sorted candidates; and outputting the sorted can

Abstract: Embodiments of the disclosure provide methods and apparatuses processing audio data. The method can include: acquiring audio data by an audio capturing device, determining feature information of an enclosure in which the audio capturing device is located, and reverberating the feature information into the audio data.

Abstract: Human-machine dialog is characterized by receiving data comprising a recording of an individual interacting with a dialog application simulating a conversation. Thereafter, the received data is parsed using automated speech recognition to result in text comprising a plurality of words. Features are extracted from the parsed data and then input an ensemble of different machine learning models each trained to generate a score characterizing a plurality of different dialog constructs. Thereafter, scores generated by the machine learning models for each of the dialog constructs are fused. A performance score is then generated based on the fused scores which characterizes a conversational proficiency of the individual interacting with the dialog application. Data can then be provided which includes or otherwise characterizes the generated score. Related apparatus, systems, techniques and articles are also described.

Abstract: An example system includes a processor to receive a data set and similarity scores. The processor is to execute an eigen response analysis on eigenvectors calculated for a similarity matrix generated based on the similarity scores for the data set. The processor is to output an estimated number of clusters in the data set based on the eigen response analysis.

Abstract: Systems and methods are described to receive a query from a user and provide a reply that is appropriate for an age group of the user. A query for a media asset is received, where such query comprises an inputted term, and the query is determined to be received from a user belonging to a first age group. A context of the inputted term within the query is identified, and in response to the determining, based on the identified context, that the inputted term of the query is inappropriate for the first age group, a replacement term for the inputted term that is related to the inputted term and is appropriate for the first age group in the context of the query is identified. The query is modified to replace the inputted term with the identified replacement term, and a reply to the modified query is generated for output.

Abstract: A processing method of a sound watermark and a speech communication system are provided. Multiple sinewave signals are generated. Frequencies of the sinewave signals are different from each other, and the sinewave signals belong to a high-frequency sound signal. A watermark pattern is mapped into a time-frequency diagram, to form a watermark sound signal. Two dimensions of the watermark pattern in a two-dimensional coordinate system respectively correspond to a time axis and a frequency axis in the time-frequency diagram. Each of multiple audio frames on the time axis corresponds to the sinewave signals with different frequencies on the frequency axis. A speech signal and the watermark sound signal are synthesized in a time domain to generate a watermark-embedded signal. Accordingly, a sound watermark may be embedded in real-time.

Abstract: An analysis filter bank corresponding to multiple sub-bands, which performs frequency-division filtering on an input signal to generate multiple sub-band signals, the analysis filter bank comprising: a sub-band response pre-compensator which performs a linear filtering on the input signal to generate a response pre-compensated signal, multiple sub-filters with different central frequencies, which perform complex-type first-order infinite impulse response filtering respectively on the response pre-compensated signal to generate multiple sub-filter signals, and multiple binomially-combining and rotating devices based on a set of binomial weights, each of which performs a weighted summation on at least two of the sub-filter signals with the set of binomial weights, and rotates a weighted-summation result with a rotating phase according to a corresponding sub-band central frequency to generate one of the sub-band signals, wherein the at least two of the sub-filter signals are generated by at least two of the sub-

Abstract: Systems and methods for contextual natural language censoring are disclosed. For example, configuration data indicating details associated with content provided by a client device and/or about the client device may be received and may be utilized to determine impermissible and permissible exceptions for a given client. One or more queries may be generated utilizing the impermissible and permissible exceptions, and when input data is received from the client device and/or in association with the client identifier, the queries may be utilized to evaluate the input data for impermissible and permissible exceptions. The results may be filtered based on user preferences, the input data may be censored, the input data may be prevented from being exposed to a user device, the application associated with the input data may be removed from availability, and/or a maturity setting may be changed for the application, for example.

Abstract: Approaches presented herein enable delivery of real-time internet of things (IoT) feedback to optimize a public speaking performance. More specifically, a set of data representing a speaking performance of a user is captured and analyzed to generate a speaking performance profile of the user. This profile is compared to a reference speaking performance profile and, based on the comparison, a set of performance improvement strategies for the user is generated. A performance improvement strategy is selected from the set of performance improvement strategies based on an identification of an availability of a set of IoT devices for delivery of at least one of the strategies. Instructions are then communicated, responsive to the captured speaking performance associated with the user, to an available IoT device to deliver the selected performance improvement strategy to the user through an output user interface of the available IoT device during the speaking performance.

Abstract: A speech translation method using a multilingual text-to-speech synthesis model includes acquiring a single artificial neural network text-to-speech synthesis model having acquired learning based on a learning text of a first language and learning speech data of the first language corresponding to the learning text of the first language, and a learning text of a second language and learning speech data of the second language corresponding to the learning text of the second language, receiving input speech data of the first language and an articulatory feature of a speaker regarding the first language, converting the input speech data of the first language into a text of the first language, converting the text of the first language into a text of the second language, and generating output speech data for the text of the second language that simulates the speaker's speech.

Abstract: A technique is provided that can reduce degradation of the sound quality due to a tandem connection of paired coding and decoding, and can reduce the operation processing amount and the required memory amount of a multipoint control unit. In multipoint connection between terminals of a plurality of communication networks (for example, a fixed phone line and a mobile phone line) having different communication capacities, when a multichannel coding including a monaural coding scheme of a communication network having a smaller communication capacity is used in a communication network having a larger communication capacity to transmit sounds of a plurality points to the communication network terminal having the larger communication capacity, control is exercised such that monaural codes of the plurality points are output.

Abstract: A system configured to perform low power mode wakeword detection is provided. A device reduces power consumption without compromising functionality by placing a primary processor into a low power mode and using a secondary processor to monitor for sound detection. The secondary processor stores input audio data in a buffer component while performing sound detection on the input audio data. If the secondary processor detects a sound, the secondary processor sends an interrupt signal to the primary processor, causing the primary processor to enter an active mode. While in the active mode, the primary processor performs wakeword detection using the buffered audio data. To reduce a latency, the primary processor processes the buffered audio data at an accelerated rate. In some examples, the device may further reduce power consumption by including a second buffer component and only processing the input audio data after detecting a sound.

Abstract: A computer-implemented method for detecting coordinated propagation of social media content may include calculating, by a computing device, a content similarity score for each social media post in relation to other social media posts in a set of social media posts. The method may also include identifying a related subset of social media posts based on the content similarity score. Additionally, the method may include detecting one or more clusters of social media posts in the related subset by clustering social media posts based on content similarity scores and timing. Furthermore, the method may include determining that a user account associated with a social media post in a detected cluster is in a coordinated network of user accounts. Finally, the method may include performing a security action in response to determining that the user account is in the coordinated network. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A computer-implemented method for analyzing call interactions in an interactions database by a Proactive Fraud Exposure (PFE) engine is provided herein. The computer-implemented method may generate a voiceprint for each call interaction; (ii) use a machine learning technique to group the call interactions into one or more clusters based on respective voiceprints in the voiceprints database; (iii) store the one or more clusters; and (iv) rank and classifying the one or more clusters to yield a list of potential fraudsters. The computer-implemented method may further transmit the list of potential fraudsters to a user to enable the user to review said list of potential fraudsters and to add fraudsters from the list to a watchlist database.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for determining a transfer option for transferring a call. One of the methods include receiving, by a call assistant engine, a keyword related to information provided by a user to an agent during a call; generating, by the call assistant engine, follow-up questions to be displayed on a user device of the agent in an interactive format, the first follow-up question being generated based on the keyword, each of the following follow-up questions being generated based on an answer of the agent to the previous question; and determining, by the call assistant engine, based on answers of the agent to the follow-up questions, a transfer option for transferring the call.