Abstract: A method of leveraging a dialogue history of a conversational computing interface to execute an updated dialogue plan. The method comprises maintaining an annotated dialogue history of the conversational computing interface. The annotated dialogue history includes a plurality of traced steps defining a data-flow including input data used to execute a context-dependent operation and output data recorded from a previous execution of the context-dependent operation. The method further comprises recognizing an updated dialogue plan including a prefix of executable steps and an updated executable step following the prefix. The method further comprises automatically computer-recognizing that the prefix of executable steps of the updated dialogue plan matches a corresponding prefix of traced steps in the annotated dialogue history. The method further comprises re-using the data-flow from the prefix of traced steps in the annotated dialogue history to automatically determine input data of the updated executable step.

Abstract: A method comprises receiving a new data set; identifying at least one prior data set of a plurality of prior data sets that matches the new data set; generating a natural language data science problem statement for the new data set based on information associated with the at least prior one data set that matches the new data set; outputting the generated natural language data science problem statement for user verification; and in response to receiving user input verifying the natural language generated data science problem statement, generating one or more AutoAI configuration settings for the new data set based on one or more AutoAI configuration settings associated with the at least one prior data set that matches the new data set.

Abstract: A method for adapting hotword recognition includes receiving audio data characterizing a hotword event detected by a first stage hotword detector in streaming audio captured by a user device. The method also includes processing, using a second stage hotword detector, the audio data to determine whether a hotword is detected by the second stage hot word detector in a first segment of the audio data. When the hotword is not detected by the second stage hotword detector, the method includes, classifying the first segment of the audio data as containing a negative hotword that caused a false detection of the hotword event in the streaming audio by the first stage hotword detector. Based on the first segment of the audio data classified as containing the negative hotword, the method includes updating the first stage hotword detector to prevent triggering the hotword event in subsequent audio data that contains the negative hotword.

Abstract: An apparatus for generating a decoded two-channel signal, comprising: a parametric decoder for providing parametric data for a second set of second spectral portions and a two-channel identification identifying for a second spectral portion of the second set of second spectral portions either a first two-channel representation for the second spectral portion of the second set of second spectral portions or a second two-channel representation for the second spectral portion of the second set of second spectral portions, the second two-channel representation being different from the first two-channel representation; and a frequency regenerator for regenerating the second spectral portion of the second set of second spectral portions depending on a first spectral portion of a first set of first spectral portions, the parametric data for the second spectral portion of the second set of second spectral portions and the two-channel identification for the second spectral portion of the second set of second spectral

Abstract: Methods, systems and apparatuses for computer-generated visualization of speech are described herein. An example method of computer-generated visualization of speech including at least one segment includes: generating a graphical representation of an object corresponding to a segment of the speech; and displaying the graphical representation of the object on a screen of a computing device. Generating the graphical representation includes: representing a duration of the respective segment by a length of the object and representing intensity of the respective segment by a width of the object; and placing, in the graphical representation, a space between adjacent objects.

Abstract: Disclosed herein are new techniques carried out by a computing system for determining delays of various components of an audio system to allow for accurate correction of these delays, which may improve the audio quality of live performances for listeners who hear audio reproduced by loudspeakers at live performance venues. In one implementation the computing system, which may comprise a transmitter device and one or more receiver devices, may be configured to perform functions, including receiving a first audio signal, receiving, via an audio input interface of the receiver, a second audio signal, and determining, based on the first audio signal and the second audio signal, an audio delay that is associated with the second audio signal. The computing system may be configured to perform further functions, including based on a determined cross-correlation between a downsampled audio signal and a filtered second audio signal, determining the audio signal delay.

Abstract: A device to perform speech enhancement includes one or more processors configured to obtain input spectral data based on an input signal. The input signal represents sound that includes speech. The one or more processors are also configured to process, using a multi-encoder transformer, the input spectral data and context data to generate output spectral data that represents a speech enhanced version of the input signal.

Abstract: Controlling a concealment method for a lost audio frame associated with a received audio signal is provided. At least one bin vector of a spectral representation for at least one tone is obtained, wherein the at least one bin vector includes three consecutive bin values for the at least one tone. Whether each of the three consecutive bin values has a complex value or a real value is determined. Responsive to the determination, the three consecutive bin values are processed to estimate a frequency of the at least one tone based on whether each bin value has a complex value or a real value.

Abstract: Technologies are disclosed for improving the efficiency of real-time audio processing, and specifically for improving the efficiency of continuously modifying a real-time audio signal. Efficiency is improved by reducing memory bandwidth requirements and by reducing the amount of processing used to modify the real-time audio signal. In some configurations, memory bandwidth requirements are reduced by selectively transferring active samples in the frequency domain—e.g. avoiding the transfer samples with amplitudes of zero or near-zero. This has particular importance when the specialized hardware retrieves samples from main memory in real-time. In some configurations, the amount of processing needed to modify the audio signal is reduced by omitting operations that do not meaningfully affect the output audio signal. For example, a multiplication of samples may be avoided when at least one of the samples has an amplitude of zero or near-zero.

Abstract: Systems and methods are described for training an artificial intelligence model to infer a log sourcetype of a log. For example, logs may have different log sourcetypes, and logs having the same log sourcetypes may have different messagetypes. The artificial intelligence model may be a machine learning model, and can be trained using training data that includes logs with known log sourcetypes. Each log can be tokenized, filtered, converted into a vector, and applied to a machine learning model as an input to perform the training. The machine learning model may output an inferred log sourcetype, which can be compared with the known log sourcetype to update model parameters to improve the machine learning model accuracy. The trained machine learning model may be trained to infer a log sourcetype of a log regardless of the messagetype of the log.

Abstract: Method for voice-based interactive communication using a digital assistant, wherein the method comprises, an attention detection step, in which the digital assistant detects a user attention and as a result is set into a listening mode; a speaker detection step, in which the digital assistant detects the user as a current speaker; a speech sound detection step, in which the digital assistant detects and records speech uttered by the current speaker, which speech sound detection step further comprises a lip movement detection step, in which the digital assistant detects a lip movement of the current speaker; a speech analysis step, in which the digital assistant parses said recorded speech and extracts speech-based verbal informational content from said recorded speech; and a subsequent response step, in which the digital assistant provides feed-back to the user based on said recorded speech.

Abstract: A computer-implemented method includes: receiving, by a computing device, an input file defining correct spellings of one or more transliterated words; generating, by the computing device, suffix outputs based on the one or more transliterated words; generating, by the computing device, a dictionary that maps the suffix outputs to the one or more transliterated words; recognizing, by the computing device, an alternatively spelled transliterated word included in a document as one of the one or more correctly spelled transliterated words using the dictionary; and outputting, by the computing device, information corresponding to the recognized transliterated word.

Abstract: A target speech signal extraction method for robust speech recognition includes: initializing a steering vector for a target speech source and an adaptive vector, setting a real output channel of the target speech source as an output by the adaptive vector, initializing adaptive vectors for a noise and setting a dummy channel as an output by the adaptive vectors for the noise; setting a cost function for minimizing dependency between a real output for the target speech source and a dummy output for the noise; setting an auxiliary function to the cost function, and updating the adaptive vector for the target speech source and the adaptive vectors for the noise by using the auxiliary function and the steering vector; estimating the target speech signal by using the adaptive vector thereby extracting the target speech signal from the input signals; and updating the steering vector for the target speech source.

Abstract: A sound signal processing system includes: a sound signal processing apparatus executing non-linear signal processing on a collected sound signal collected by a microphone, and transmitting, to an information processing apparatus, both a pre-execution sound signal before the non-linear signal processing is executed and a post-execution sound signal after the non-linear signal processing is executed; and the information processing apparatus receiving the pre-execution sound signal and the post-execution sound signal from the sound signal processing apparatus, and executing first processing on the pre-execution sound signal and executing second processing on the post-execution sound signal, the second processing being different from the first processing.

Abstract: Implementations relate to generating and/or executing a customized interactive dialog application. The customized interactive dialog application may be generated from a state mapping tool that allows a user to generate custom states and custom transitions between the custom states. A customized configuration description is then generated based on the generated custom states and custom transitions. Further, a default configuration description is identified that includes additional or alternative states and transitions. In executing the customized interactive dialog application, dialog turns are generated based on the states and transition information, with the customized configuration description taking precedence and the default configuration description being utilized for any undefined states and/or transitions. Implementations additionally or alternatively relate to generating and/or executing a custom agent based on generated custom states and custom transitions, and a default configuration description.

Abstract: An apparatus for processing an audio signal includes an audio signal analyzer and a filter. The audio signal analyzer is configured to analyze an audio signal to determine a plurality of noise suppression filter values for a plurality of bands of the audio signal, wherein the analyzer is configured to determine a noise suppression filter value so that a noise suppression filter value is greater than or equal to a minimum noise suppression filter value and so that the minimum noise suppression value depends on a characteristic of the audio signal. The filter is configured for filtering the audio signal, wherein the filter is adjusted based on the noise suppression filter values.

Abstract: A method of biasing speech recognition includes receiving audio data encoding an utterance and obtaining a set of one or more biasing phrases corresponding to a context of the utterance. Each biasing phrase in the set of one or more biasing phrases includes one or more words. The method also includes processing, using a speech recognition model, acoustic features derived from the audio data and grapheme and phoneme data derived from the set of one or more biasing phrases to generate an output of the speech recognition model. The method also includes determining a transcription for the utterance based on the output of the speech recognition model.

Abstract: The present document relates to audio source coding systems which make use of a harmonic transposition method for high frequency reconstruction (HFR), as well as to digital effect processors, e.g. exciters, where generation of harmonic distortion add brightness to the processed signal, and to time stretchers where a signal duration is prolonged with maintained spectral content. A system and method configured to generate a time stretched and/or frequency transposed signal from an input signal is described. The system comprises an analysis filterbank configured to provide an analysis subband signal from the input signal; wherein the analysis subband signal comprises a plurality of complex valued analysis samples, each having a phase and a magnitude. Furthermore, the system comprises a subband processing unit configured to determine a synthesis subband signal from the analysis subband signal using a subband transposition factor Q and a subband stretch factor S.

Abstract: A language proficiency analyzer automatically evaluates a person's language proficiency by analyzing that person's oral communications with another person. The analyzer first enhances the quality of an audio recording of a conversation between the two people using a neural network that automatically detects loss features in the audio and adds those loss features back into the audio. The analyzer then performs a textual and audio analysis on the improved audio. Through textual analysis, the analyzer uses a multi-attention network to determine how focused one person is on the other and/or how pleased one person is with the other. Through audio analysis, the analyzer uses a neural network to determine how well one person pronounced words during the conversation.

Abstract: A method of operating a customer utterance analysis system includes obtaining a subset of utterances from among a first set of utterances. The method includes encoding, by a sentence encoder, the subset of utterances into multi-dimensional vectors. The method includes generating reduced-dimensionality vectors by reducing a dimensionality of the multi-dimensional vectors. Each vector of the reduced-dimensionality vectors corresponds to an utterance from among the subset of utterances. The method includes performing clustering on the reduced-dimensionality vectors. The method includes, based on the clustering performed on the reduced-dimensionality vectors, arranging the subset of utterances into clusters. The method includes obtaining labels for a least two clusters from among the clusters. The method includes generating training data based on the obtained labels. The method includes training a neural network model to predict an intent of an utterance based on the training data.