Abstract: Without dividing speech into a unit such as a word or a character, text corresponding to the speech is labeled. A speech distributed representation sequence converting unit 11 converts an acoustic feature sequence into a speech distributed representation. A symbol distributed representation converting unit 12 converts each symbol included in the symbol sequence corresponding to the acoustic feature sequence into a symbol distributed representation. A label estimation unit 13 estimates a label corresponding to the symbol from the fixed-length vector of the symbol generated using the speech distributed representation, the symbol distributed representation, and fixed-length vectors of previous and next symbols.

Abstract: Methods, apparatus, and systems are disclosed to segment audio and determine audio segment similarities. An example apparatus includes at least one memory storing instructions and processor circuitry to execute instructions to at least select an anchor index beat of digital audio, identify a first segment of the digital audio based on the anchor index beat to analyze, the first segment having at least two beats and a respective center beat, concatenate time-frequency data of the at least two beats and the respective center beat to form a matrix of the first segment, generate a first deep feature based on the first segment, the first deep feature indicative of a descriptor of the digital audio, and train internal coefficients to classify the first deep feature as similar to a second deep feature based on the descriptor of the first deep feature and a descriptor of a second deep feature.

Abstract: A method for performing speech recognition using sequence-to-sequence models includes receiving audio data for an utterance and providing features indicative of acoustic characteristics of the utterance as input to an encoder. The method also includes processing an output of the encoder using an attender to generate a context vector, generating speech recognition scores using the context vector and a decoder trained using a training process, and generating a transcription for the utterance using word elements selected based on the speech recognition scores. The transcription is provided as an output of the ASR system.

Abstract: An apparatus for processing speech data may include a processor configured to: separate an input speech into speech signals; identify a bandwidth of each of the speech signals; extract speaker embeddings from the speech signals based on the bandwidth of each of the speech signals, using at least one neural network configured to receive the speech signals and output the speaker embeddings; and cluster the speaker embeddings into one or more speaker clusters, each speaker cluster corresponding to a speaker identity.

Abstract: In one embodiment, a method includes receiving a user's utterance comprising a word in a custom vocabulary list of the user, generating a previous token to represent a previous audio portion of the utterance, and generating a current token to represent a current audio portion of the utterance by generating a bias embedding by using the previous token to query a trie of wordpieces representing the custom vocabulary list, generating first probabilities of respective first candidate tokens likely uttered in the current audio portion based on the bias embedding and the current audio portion, generating second probabilities of respective second candidate tokens likely uttered after the previous token based on the previous token and the bias embedding, and generating the current token to represent the current audio portion of the utterance based on the first probabilities of the first candidate tokens and the second probabilities of the second candidate tokens.

Abstract: Representative embodiments disclose machine learning classifiers used in scenarios such as speech recognition, image captioning, machine translation, or other sequence-to-sequence embodiments. The machine learning classifiers have a plurality of time layers, each layer having a time processing block and a depth processing block. The time processing block is a recurrent neural network such as a Long Short Term Memory (LSTM) network. The depth processing blocks can be an LSTM network, a gated Deep Neural Network (DNN) or a maxout DNN. The depth processing blocks account for the hidden states of each time layer and uses summarized layer information for final input signal feature classification. An attention layer can also be used between the top depth processing block and the output layer.

Abstract: This specification relates to the generation of animation data using recurrent neural networks. According to a first aspect of this specification, there is described a computer implemented method comprising: sampling an initial hidden state of a recurrent neural network (RNN) from a distribution; generating, using the RNN, a sequence of frames of animation from the initial state of the RNN and an initial set of animation data comprising a known initial frame of animation, the generating comprising, for each generated frame of animation in the sequence of frames of animation: inputting, into the RNN, a respective set of animation data comprising the previous frame of animation data in the sequence of frames of animation; generating, using the RNN and based on a current hidden state of the RNN, the frame of animation data; and updating the hidden state of the RNN based on the input respective set of animation data.

Abstract: The present disclosure relates to a speech enhancement apparatus, and specifically, to a method and apparatus for a target exaggeration for deep learning-based speech enhancement. According to an embodiment of the present disclosure, the apparatus for a target exaggeration for deep learning-based speech enhancement can preserve a speech signal from a noisy speech signal and can perform speech enhancement for removing a noise signal.

Abstract: Methods, systems, and apparatuses for predicting an end of a command in a voice recognition input are described herein. The system may receive data comprising a voice input. The system may receive a signal comprising a voice input. The system may detect, in the voice input, data that is associated with a first portion of a command. The system may predict, based on the first portion and while the voice input is being received, a second portion of the command. The prediction may be generated by a machine learning algorithm that is trained based at least in part on historical data comprising user input data. The system may cause execution of the command, based on the first portion and the predicted second portion, prior to an end of the voice input.

Abstract: A recurrent neural network-transducer (RNN-T) model improves speech recognition by processing sequential non-blank symbols at each time step after an initial one. The model's prediction network receives a sequence of symbols from a final Softmax layer and employs a shared embedding matrix to create and map embeddings to each symbol, associating them with unique position vectors. These embeddings are weighted according to their similarity to their matching position vector. Subsequently, a joint network of the RNN-T model uses these weighted embeddings to output a probability distribution for potential speech recognition hypotheses at each time step, enabling more accurate transcriptions of spoken language.

Abstract: A system and method for publishing encoded identity data that uses at least biometric information as well as non-biometric identity and/or authentication data is disclosed. The system and method can be used for verifying a user's identity against the published encoded identity data on a distributed system, such as a distributed ledger or blockchain. Using this system, a user's identity can be verified efficiently by multiple parties, in sequence, or in parallel, as a user need only enroll in the verification process a single time. The system further includes a biometric enrollment sub-system that allows for a highly secure method of verifying a user based on unique biometric signals, such as features extracted from an audio voice signal.

Abstract: Embodiments are disclosed for noise floor estimation and noise reduction, In an embodiment, a method comprises: obtaining an audio signal; dividing the audio signal into a plurality of buffers; determining time-frequency samples for each buffer of the audio signal; for each buffer and for each frequency, determining a median (or mean) and a measure of an amount of variation of energy based on the samples in the buffer and samples in neighboring buffers that together span a specified time range of the audio signal; combining the median (or mean) and the measure of the amount of variation of energy into a cost function; for each frequency: determining a signal energy of a particular buffer of the audio signal that corresponds to a minimum value of the cost function; selecting the signal energy as the estimated noise floor of the audio signal; and reducing, using the estimated noise floor, noise in the audio signal.

Abstract: Systems and methods for facilitating a watch party are provided. In one example, a method includes: initiating a watch party session for a host user using, presenting content selected by the host user on a first user device during the watch party session, initiating a chat session concurrent with the watch party session, receiving a participation request by a guest user sent from a second user device for participating in the chat session; in response to the participation request, authenticating the guest user; presenting the content selected by the host user on the second user device; synchronizing the presentation of the content on the first user device with the presentation of the second user device; and facilitating communication between the host user and the guest user during the chat session.

Abstract: A library of machine learning primitives is provided to optimize a machine learning model to improve the efficiency of inference operations. In one embodiment a trained convolutional neural network (CNN) model is processed into a trained CNN model via pruning, convolution window optimization, and quantization.

Abstract: An audio keyword searcher arranged to identify a voice segment of a received audio signal; identify, by an automatic speech recognition engine, one or more phonemes included in the voice segment; output, from the automatic speech recognition engine, the one or more phonemes to a keyword filter to detect whether the voice segment includes any of the one or more first keywords of the first keyword list and, if detected, output the one or more phonemes included in the voice segment to a decoder but, if not detected, not output the one or more phonemes included in the voice segment to the decoder. If the one or more phonemes are output to the decoder: generate a word lattice associated with the voice segment; search the word lattice for one or more second keywords, and determine whether the voice segment includes the one or more second keywords.

Abstract: This application relates to apparatus and methods for generating preference profiles that may be used to rank search results. In some examples, a computing device obtains browsing session data and determines items that were engaged, such as items that were viewed or clicked. The computing device obtains item property data, such as product descriptions, for the items, and applies a dependency parser to the item property data to identify portions that include certain words, such as nouns or adjectives, which are then identified as attributes. The computing device generates attribute data identifying portions of the item property data as item attributes. In some examples, the computing device applies one or more machine learning algorithms to the session data and/or search query to identify item attributes. The computing device may generate a profile that includes the item attributes, and may rank search results based on the attribute data, among other uses.

Abstract: Systems and techniques are provided for processing one or more data samples. For example, a neural network classifier can be trained to perform few-shot open-set recognition (FSOSR) based on a task-agnostic open-set prototype. A process can include determining one or more prototype representations for each class included in a plurality of support samples. A task-agnostic open-set prototype representation can be determined, in a same learned metric space as the one or more prototype representations. One or more distance metrics can be determined for each query sample of one or more query samples, based on the one or more prototype representations and the task-agnostic open-set prototype representation. Based on the one or more distance metrics, each query sample can be classified into one of classes associated with the one or more prototype representations or an open-set class associated with the task-agnostic open-set prototype representation.

Abstract: This document relates to architectures and training procedures for multi-task machine learning models, such as neural networks. One example method involves providing a multi-task machine learning model having one or more shared layers and two or more task-specific layers. The method can also involve performing a pretraining stage on the one or more shared layers using one or more unsupervised prediction tasks. The method can also involve performing a tuning stage on the one or more shared layers and the two or more task-specific layers using respective task-specific objectives.

Abstract: An embodiment dialogue system includes a speech recognizer configured to convert an utterance of a user into an utterance text, a natural language understanding module configured to identify an intention of the user based on the utterance text, and a controller configured to generate a first control signal for performing control corresponding to the intention of the user, identify whether an additional control item related to the control corresponding to the intention of the user exists, and in response to the additional control item existing, generate a second control signal for displaying information about the additional control item on a display.

Abstract: Embodiments described a method of video-text pre-learning to effectively learn cross-modal representations from sparse video frames and text. Specifically, an align and prompt framework provides a video and language pre-training framework that encodes the frames and text independently using a transformer-based video encoder and a text encoder. A multi-modal encoder is then employed to capture cross-modal interaction between a plurality of video frames and a plurality of texts. The pre-training includes a prompting entity modeling that enables the model to capture fine-grained region-entity alignment.

Abstract: The present technology relates to a particular-sound detector and method, and a program that make it possible to improve the performance of detecting particular sounds. The particular-sound detector includes a particular-sound detecting section that detects a particular sound on a basis of a plurality of audio signals obtained by collecting sounds by a plurality of microphones provided to a wearable device. In addition, the plurality of the microphones includes two microphones that are equidistant at least from a sound source of the particular sound, and one microphone arranged at a predetermined position. The present technology can be applied to headphones.

Abstract: The invention relates to a system for interacting with an occupant of a motor vehicle comprising: a. a measuring device comprising at least one sensor arranged to acquire at least one parameter associated with the occupant of said vehicle; b. an on-board processing unit arranged to receive said parameter and to define a data item representing the emotional state of said occupant by means of said model, said representative data item being a comfort index score (CISn) of said occupant; c. the representative data item corresponding to a point in a two-dimensional space (anvn) for characterising the emotional state of the occupant; d. characterised in that an emotional comfort index is computed on the basis of the representative data item; e. and in that at least one actuator is configured to activate at least one multi-sensory stimulus for interacting with the occupant, said stimulus allowing the emotional state of said occupant to be changed.

Abstract: Disclosed is a method for detecting a voice from audio data, performed by a computing device according to an exemplary embodiment of the present disclosure. The method includes obtaining audio data; generating image data based on a spectrum of the obtained audio data; analyzing the generated image data by utilizing a pre-trained neural network model; and determining whether an automated response system (ARS) voice is included in the audio data, based on the analysis of the image data.

Abstract: A method includes receiving a plurality of unlabeled audio samples corresponding to spoken utterances not paired with corresponding transcriptions. At a target branch of a contrastive Siamese network, the method also includes generating a sequence of encoder outputs for the plurality of unlabeled audio samples and modifying time characteristics of the encoder outputs to generate a sequence of target branch outputs. At an augmentation branch of a contrastive Siamese network, the method also includes performing augmentation on the unlabeled audio samples, generating a sequence of augmented encoder outputs for the augmented unlabeled audio samples, and generating predictions of the sequence of target branch outputs generated at the target branch. The method also includes determining an unsupervised loss term based on target branch outputs and predictions of the sequence of target branch outputs. The method also includes updating parameters of the audio encoder based on the unsupervised loss term.

Abstract: Embodiments described herein propose a densely connected Transformer architecture in which each Transformer layer takes advantages of all previous layers. Specifically, the input for each Transformer layer comes from the outputs of all its preceding layers; and the output information of each layer will be incorporated in all its subsequent layers. In this way, a L-layer Transformer network will have L(L+1)/2 connections. In this way, the dense connection allows the linguistic information learned by the lower layer to be directly propagated to all upper layers and encourages feature reuse throughout the network. Each layer is thus directly optimized from the loss function in the fashion of implicit deep supervision.

Abstract: A speaker verification method includes receiving audio data corresponding to an utterance, processing a first portion of the audio data that characterizes a predetermined hotword to generate a text-dependent evaluation vector, and generating one or more text-dependent confidence scores. When one of the text-dependent confidence scores satisfies a threshold, the operations include identifying a speaker of the utterance as a respective enrolled user associated with the text-dependent confidence score that satisfies the threshold and initiating performance of an action without performing speaker verification. When none of the text-dependent confidence scores satisfy the threshold, the operations include processing a second portion of the audio data that characterizes a query to generate a text-independent evaluation vector, generating one or more text-independent confidence scores, and determining whether the identity of the speaker of the utterance includes any of the enrolled users.

Abstract: Examples are provided relating to recovering depth data from noisy phase data of low-signal pixels. One example provides a computing system, comprising a logic machine, and a storage machine holding instructions executable by the logic machine to process depth data by obtaining depth image data and active brightness image data for a plurality of pixels, the depth image data comprising phase data for a plurality of frequencies, and identifying low-signal pixels based at least on the active brightness image data. The instructions are further executable to apply a denoising filter to phase data of the low-signal pixels to obtain denoised phase data and not applying the denoising filter to phase data of other pixels. The instructions are further executable to, after applying the denoising filter, perform phase unwrapping on the phase data for the plurality of frequencies to obtain a depth image, and output the depth image.

Abstract: A supervised discriminator for detecting bio-markers in an audio sample dataset is trained and a denoising autoencoder is trained to learn a latent space that is used to reconstruct an output audio sample with a same fidelity as an input audio sample of the audio sample dataset. A conditional auxiliary generative adversarial network (GAN) trained to generate the output audio sample with the same fidelity as the input audio sample, wherein the output audio sample is void of the bio-markers. The conditional auxiliary generative adversarial network (GAN), the corresponding supervised discriminator, and the corresponding denoising autoencoder are deployed in an audio processing system.

Abstract: Methods, apparatus, systems, and articles of manufacture to load data into an accelerator are disclosed. An example apparatus includes data provider circuitry to load a first section and an additional amount of compressed machine learning parameter data into a processor engine. Processor engine circuitry executes a machine learning operation using the first section of compressed machine learning parameter data. A compressed local data re-user circuitry determines if a second section is present in the additional amount of compressed machine learning parameter data. The processor engine circuitry executes a machine learning operation using the second section when the second section is present in the additional amount of compressed machine learning parameter data.

Abstract: A computer implemented method of storing and retrieving feature map data of a neural network the method comprising receiving a first portion of feature map data from local storage, selecting a first set of subportions of the first portion of feature map data, compressing the subportions to produce a first plurality of sections of compressed feature map data and instructing the storage of the sections into external storage. The method also comprises receiving a second plurality of sections of compressed feature map data from the external storage, decompressing the sections to produce a second set of subportions of the second portion of feature map data and storing the second portion of feature map data in local storage. The first and second sets of subportions each correspond to a predetermined format of subdivision and the method comprises selecting the predetermined format of subdivision from a plurality of predetermined formats of subdivision.

Abstract: According to an aspect, method for synthesizing multi-speaker speech using an artificial neural network comprises generating and storing a speech learning model for a plurality of users by subjecting a synthetic artificial neural network of a speech synthesis model to learning, based on speech data of the plurality of users, generating speaker vectors for a new user who has not been learned and the plurality of users who have already been learned by using a speaker recognition model, determining a speaker vector having the most similar relationship with the speaker vector of the new user according to preset criteria out of the speaker vectors of the plurality of users who have already been learned, and generating and learning a speaker embedding of the new user by subjecting the synthetic artificial neural network of the speech synthesis model to learning, by using a value of a speaker embedding of a user for the determined speaker vector as an initial value and based on speaker data of the new user.

Abstract: Embodiments are disclosed for real time generative audio for brush and canvas interaction in digital drawing. The method may include receiving a user input and a selection of a tool for generating audio for a digital drawing interaction. The method may further include generating intermediary audio data based on the user input and the tool selection, wherein the intermediary audio data includes a pitch and a frequency. The method may further include processing, by a trained audio transformation model and through a series of one or more layers of the trained audio transformation model, the intermediary audio data. The method may further include adjusting the series of one or more layers of the trained audio transformation model to include one or more additional layers to produce an adjusted audio transformation model. The method may further include generating, by the adjusted audio transformation model, an audio sample based on the intermediary audio data.

Abstract: Introduced here is a computer program that is representative of a software-implemented collaboration platform that is designed to facilitate conversations in virtual environments, document those conversations, and analyze those conversations, all in real time. The collaboration platform can include or integrate tools for turning ideas—expressed through voice—into templatized, metadata-rich data structures called “knowledge objects.” Discourse throughout a conversation can be converted into a transcription (or simply “transcript”), parsed to identify topical shifts, and then segmented based on the topical shifts. Separately documenting each topic in the form of its own “knowledge object” allows the collaboration platform to not only better catalogue what was discussed in a single ideation session, but also monitor discussion of the same topic over multiple ideation sessions.

Abstract: The present disclosure provides methods and systems for providing machine learning model service. The method may comprise: (a) generating, by a first computing system, a first output data using a first machine learning model, wherein the first machine learning model is trained on a first training dataset; (b) transmitting the first output data to a second computing system, wherein the first training dataset and the first machine learning model are inaccessible to the second computing system; (c) creating an input data by joining the first output data with a selected set of input features accessible to the second computing system; and (d) generating a second output data using a second machine learning model to process the input data.

Abstract: Audio signals of speech may be processed using an acoustic model. An acoustic model may be implemented with multiple streams of processing where different streams perform processing using different dilation rates. For example, a first stream may process features of the audio signal with one or more convolutional neural network layers having a first dilation rate, and a second stream may process features of the audio signal with one or more convolutional neural network layers having a second dilation rate. Each stream may compute a stream vector, and the stream vectors may be combined to a vector of speech unit scores, where the vector of speech unit scores provides information about the acoustic content of the audio signal. The vector of speech unit scores may be used for any appropriate application of speech, such as automatic speech recognition.

Abstract: An apparatus and method for enhancing broadcast radio includes a DNN trained on data sets of audio created from a synthesized broadcasting process and original source audio. Broadcast radio signals are received at a radio module and processed through the DNN to produce enhanced audio.

Abstract: Provided is an in-ear device and associated computational support system that leverages machine learning to interpret sensor data descriptive of one or more in-ear phenomena during subvocalization by the user. An electronic device can receive sensor data generated by at least one sensor at least partially positioned within an ear of a user, wherein the sensor data was generated by the at least one sensor concurrently with the user subvocalizing a subvocalized utterance. The electronic device can then process the sensor data with a machine-learned subvocalization interpretation model to generate an interpretation of the subvocalized utterance as an output of the machine-learned subvocalization interpretation model.

Abstract: Disclosed is a computer implemented method for malware detection that analyses a file on a per packet basis. The method receives a packet of one or more packets associated a file, and converting a binary content associated with the packet into a digital representation and tokenizing plain text content associated with the packet. The method extracts one or more n-gram features, an entropy feature, and a domain feature from the converted content of the packet and applies a trained machine learning model to the one or more features extracted from the packet. The output of the machine learning method is a probability of maliciousness associated with the received packet. If the probability of maliciousness is above a threshold value, the method determines that the file associated with the received packet is malicious.

Abstract: A learning device calculates an image feature using a model (image encoder) that receives an image and outputs the image feature obtained by mapping the image into a latent space. The learning device calculates an audio feature using a model (audio encoder) that receives a speech in a predetermined language and outputs the audio feature obtained by mapping the speech into the latent space, and that includes a neural network provided with a self-attention mechanism. The learning device updates parameters of the models used by an image feature calculation unit and an audio feature calculation unit such that the image feature of a first image is similar to the audio feature of a speech corresponding to the first image.

Abstract: A computer system analyses audio data representing a user speaking words from a body of text and identifies occasions where the user mispronounces an expected phoneme. Mispronunciation of the expected phoneme is identified by comparison with a phonetic sequence corresponding to the text, based on a predetermined or user-selected language model. The system requires the user to read continuously for a period of time, so that the user cannot hide any tendency they have to pronounce the words of the text either incorrectly or differently to the expected phonemes from the language model. The system operates on the basis of comparing the similarity of the spoken sounds of the user with the expected phonemes for the body of text, and it is not necessary to convert the user's speech to text. As the computer system need only work with the similarity scores and the sequence of expected phonemes, it can be implemented in a computationally efficient manner.

Abstract: The present disclosure provides an artificial intelligence (AI) system for sequence-to-sequence modeling with attention adapted for streaming applications. The AI system comprises at least one processor; and memory having instructions stored thereon that, when executed by the processor, cause the AI system to process each input frame in a sequence of input frames through layers of a deep neural network (DNN) to produce a sequence of outputs. At least some of the layers of the DNN include a dual self-attention module having a dual non-causal and causal architecture attending to non-causal frames and causal frames. Further, the AI system renders the sequence of outputs.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for neural programming. One of the methods includes processing a current neural network input using a core recurrent neural network to generate a neural network output; determining, from the neural network output, whether or not to end a currently invoked program and to return to a calling program from the set of programs; determining, from the neural network output, a next program to be called; determining, from the neural network output, contents of arguments to the next program to be called; receiving a representation of a current state of the environment; and generating a next neural network input from an embedding for the next program to be called and the representation of the current state of the environment.

Abstract: Systems and methods for generating training data for a supervised topic modeling system from outputs of a topic discovery model are described herein. In an embodiment, a system receives a plurality of digitally stored call transcripts and, using a topic model, generates an output which identifies a plurality of topics represented in the plurality of digitally stored call transcripts. Using the output of the topic model, the system generates an input dataset for a supervised learning model by identify a first subset of the plurality of digitally stored call transcripts that include the particular topic, storing a positive value for the first subset, identifying a second subset that do not include the particular topic, and storing a negative value for the second subset. The input training dataset is then used to train a supervised learning model.

Abstract: Systems, methods, and non-transitory computer-readable media can obtain a stream of audio waveform data that represents speech involving a plurality of speakers. As the stream of audio waveform data is obtained, a plurality of audio chunks can be determined. An audio chunk can be associated with one or more identity embeddings. The stream of audio waveform data can be segmented into a plurality of segments based on the plurality of audio chunks and respective identity embeddings associated with the plurality of audio chunks. A segment can be associated with a speaker included in the plurality of speakers. Information describing the plurality of segments associated with the stream of audio waveform data can be provided.

Abstract: A system uses conversation engines to process natural language requests and conduct automatic conversations with users. The system generates responses to users in an online conversation. The system ranks generated user responses for the online conversation. The system generates a context vector based on a sequence of utterances of the conversation and generates response vectors for generated user responses. The system ranks the user responses based on a comparison of the context vectors and user response vectors. The system uses a machine learning based model that uses a pretrained neural network that supports multiple languages. The system determines a context of an utterance based on utterances in the conversation. The system generates responses and ranks them based on the context. The ranked responses are used to respond to the user.

Abstract: A computer-implemented method is provided for model training. The method includes training a second end-to-end neural speech recognition model that has a bidirectional encoder to output same symbols from an output probability lattice of the second end-to-end neural speech recognition model as from an output probability lattice of a trained first end-to-end neural speech recognition model having a unidirectional encoder. The method also includes building a third end-to-end neural speech recognition model that has a unidirectional encoder by training the third end-to-end neural speech recognition model as a student by using the trained second end-to-end neural speech recognition model as a teacher in a knowledge distillation method.

Abstract: A processing unit can train a model as a joint multi-domain recurrent neural network (JRNN), such as a bi-directional recurrent neural network (bRNN) and/or a recurrent neural network with long-short term memory (RNN-LSTM) for spoken language understanding (SLU). The processing unit can use the trained model to, e.g., jointly model slot filling, intent determination, and domain classification. The joint multi-domain model described herein can estimate a complete semantic frame per query, and the joint multi-domain model enables multi-task deep learning leveraging the data from multiple domains. The joint multi-domain recurrent neural network (JRNN) can leverage semantic intents (such as, finding or identifying, e.g., a domain specific goal) and slots (such as, dates, times, locations, subjects, etc.) across multiple domains.

Abstract: A neural network mapping method and a neural network mapping apparatus are provided. The method includes: mapping a calculation task for a preset feature map of each network layer in a plurality of network layers in a convolutional neural network to at least one processing element of a chip; acquiring the number of phases needed by a plurality of processing elements in the chip for completing the calculation tasks, and performing a first stage of balancing on the number of phases of the plurality of processing elements; and based on the number of the phases of the plurality of processing elements obtained after the first stage of balancing, mapping the calculation task for the preset feature map of each network layer in the plurality of network layers in the convolutional neural network to at least one processing element of the chip subjected to the first stage of balancing.

Abstract: Techniques are described automatically determining runtime configurations used to execute recurrent neural networks (RNNs) for training or inference. One such configuration involves determining whether to execute an RNN in a looped, or “rolled,” execution pattern or in a non-looped, or “unrolled,” execution pattern. Execution of an RNN using a rolled execution pattern generally consumes less memory resources than execution using an unrolled execution pattern, whereas execution of an RNN using an unrolled execution pattern typically executes faster. The configuration choice thus involves a time-memory tradeoff that can significantly affect the performance of the RNN execution. This determination is made automatically by a machine learning (ML) runtime by analyzing various factors such as, for example, a type of RNN being executed, the network structure of the RNN, characteristics of the input data to the RNN, an amount of computing resources available, and so forth.

Abstract: A hearing aid with the variable number of channels includes: a microphone that receives a sound signal; an AD converter that converts the sound signal input from the microphone into a digital signal and outputs the converted digital signal; a controller that determines a filter bank channel for processing the digital signal output from the AD converter; a buffer unit that delays the digital signal based on the determined filter bank channel; a signal processor that includes at least one filter bank channel, synthesizes the digital signal using the determined filter bank channel and outputs the synthesized digital signal; a DA converter that converts the digital signal into the sound signal and outputs the converted sound signal; and a speaker that outputs the sound signal output from the DA converter, in which the controller determines the filter bank channel for processing the digital signal based on a preset condition.