Abstract: A method of attention-based end-to-end (A-E2E) automatic speech recognition (ASR) training, includes performing cross-entropy training of a model, based on one or more input features of a speech signal, determining a posterior probability vector at a time of a first wrong token among one or more output tokens of the model of which the cross-entropy training is performed, and determining a loss of the first wrong token at the time, based on the determined posterior probability vector. The method further includes determining a total loss of a training set of the model of which the cross-entropy training is performed, based on the determined loss of the first wrong token, and updating the model of which the cross-entropy training is performed, based on the determined total loss of the training set.

Abstract: A method for generating an automatic speech recognition (ASR) model using unsupervised learning includes obtaining, by a device, text information. The method includes determining, by the device, a set of phoneme sequences associated with the text information. The method includes obtaining, by the device, speech waveform data. The method includes determining, by the device, a set of phoneme boundaries associated with the speech waveform data. The method includes generating, by the device, the ASR model using an output distribution matching (ODM) technique based on determining the set of phoneme sequences associated with the text information and based on determining the set of phoneme boundaries associated with the speech waveform data.

Abstract: A method of performing speech synthesis, includes encoding character embeddings, using any one or any combination of convolutional neural networks (CNNs) and recurrent neural networks (RNNs), applying a relative-position-aware self attention function to each of the character embeddings and an input mel-scale spectrogram, and encoding the character embeddings to which the relative-position-aware self attention function is applied. The method further includes concatenating the encoded character embeddings and the encoded character embeddings to which the relative-position-aware self attention function is applied, to generate an encoder output, applying a multi-head attention function to the encoder output and the input mel-scale spectrogram to which the relative-position-aware self attention function is applied, and predicting an output mel-scale spectrogram, based on the encoder output and the input mel-scale spectrogram to which the multi-head attention function is applied.

Abstract: Disclosed is a mobile device and a control method thereof. The mobile device includes a communication interface configured to perform communication via a network; a display configured to display an image; and a processor.

Abstract: A first electronic device is provided that detects a second electronic device disposed in an area related to the first electronic device while obtaining video. Based on detecting the second electronic device disposed in the area, the video is automatically transmitted, without receiving a user input for transmitting the video, to the second electronic device such that the transmitted video is provided on a display of the second electronic device. A third electronic device disposed in the area related to the first electronic device is detected while providing a still image on the display of the first electronic device. Based on detecting the third electronic device disposed in the area, the still image is automatically transmitted, without receiving a user input for transmitting the still image, to the third electronic device such that the transmitted still image is provided on a display of the third electronic device.

Abstract: An example method of notifying a user of a message received by a first device of the user includes connecting, by the first device, to a second device of the user through a local communication network, wherein the second device is currently registered by the first device; receiving, by the first device, a message, from an external device of another user; in response to determining the first device is not currently being used when the message is received, determining whether the registered second device is currently being used; and transmitting notification information of the received message to the second device currently being used by the user, wherein a notification regarding the received message is displayed on a display of the second device currently being used by the user.

Abstract: Nucleic acid based vaccine constructs encoding Lyssaviral antigens are useful in preventing and treating diseases. Self-amplifying RNA molecules encoding Lyssaviral antigens provide potent and long-lasting immunity.

Abstract: Provided are a speech recognition training processing method and an apparatus including the same. The speech recognition training processing method includes acquiring a stream of speech data from one or more speakers, extracting an auxiliary feature corresponding to a speech characteristic of the one or more speaker and updating an acoustic model by performing permutation invariant training (PIT) model training based on the auxiliary feature.

Abstract: Provided are a speech recognition training processing method and an apparatus including the same. The speech recognition training processing method includes acquiring multi-talker mixed speech sequence data corresponding to a plurality of speakers, encoding the multi-speaker mixed speech sequence data into an embedded sequence data, generating speaker specific context vectors at each frame based on the embedded sequence, generating senone posteriors for each of the speaker based on the speaker specific context vectors and updating an acoustic model by performing permutation invariant training (PIT) model training based on the senone posteriors.

Abstract: Provided are a speech recognition training processing method and an apparatus including the same. The speech recognition training processing method includes acquiring a multi-talker mixed speech signal from a plurality of speakers, performing permutation invariant training (PIT) model training on the multi-talker mixed speech signal based on knowledge from a single-talker speech recognition model and updating a multi-talker speech recognition model based on a result of the PIT model training.

Abstract: An example system includes an electronic device and a server. The electronic device includes a display; a first communicator; and a first controller configured to: control to transmit, to the server, location data identifying locations of the electronic device. The server includes a second communicator; and a second controller configured to: control to receive the location data; identify a course of travel; link photograph images to locations on the course of travel; and control to transmit to the electronic device the course of travel and the photograph images. The first controller of the electronic device is configured to control to receive the course of travel and the photograph images and to control to display on the display, at the same time, a map image, the course of travel, and one or more of the photograph images linked to a selected location on the course of travel.

Abstract: Systems and methods for providing a predictive framework are provided. The predictive framework comprises plural neural layers of adaptable, executable neurons. Neurons accept one or more input signals and produce an output signal that may be used by an upper-level neural layer. Input signals are received by an encoding neural layer, where there is a 1:1 correspondence between an input signal and an encoding neuron. Input signals for a set of data are received at the encoding layer and processed successively by the plurality of neural layers. An objective function utilizes the output signals of the topmost neural layer to generate predictive results for the data set according to an objective. In one embodiment, the objective is to determine the likelihood of user interaction with regard to a specific item of content in a set of search results, or the likelihood of user interaction with regard to any item of content in a set of search results.

Abstract: Methods and apparatuses are provided for performing end-to-end speech recognition training performed by at least one processor. The method includes receiving, by the at least one processor, one or more input speech frames, generating, by the at least one processor, a sequence of encoder hidden states by transforming the input speech frames, computing, by the at least one processor, attention weights based on each of the sequence of encoder hidden states and a current decoder hidden state, performing, by the at least one processor, a decoding operation based on a previous embedded label prediction information and a previous attentional hidden state information generated based on the attention weights; and generating a current embedded label prediction information based on a result of the decoding operation and the attention weights.

Abstract: Disclosed is a mobile device, a control method thereof, a message sending apparatus and a message sending method. The mobile device includes a wireless communication unit which performs wireless communication via a wireless network; a signal processing unit which processes a signal transmitted/received by the wireless communication unit; a display unit which displays an image based on the signal processed by the signal processing unit; and a control unit which receives a message from a counterpart device, displays the received message on the display unit, remotely accesses a remote device, and controls the remote device to allow the mobile device to access information stored in the remote device.

Abstract: Provided is a system and method for increasing file transmission speed of a server, and a method for replicating a data object over a plurality of nodes. The system includes a server communicatively coupled to a plurality of nodes. The server may detect a plurality of nodes. The server may determine that a data object stored by the server is to be replicated to the plurality of nodes. The server may divide the data object into a plurality of data pieces. The server may send a different data piece to each node of the plurality of nodes. Each node may receive the different data piece from the server. Each node may send a copy of the received data piece to all other nodes. Each node may receive copies of all other data pieces from the other nodes. Each node may replicate a full copy of the data object.

Abstract: A method and apparatus are provided that analyzing sequence-to-sequence data, such as sequence-to-sequence speech data or sequence-to-sequence machine translation data for example, by minimum Bayes risk (MBR) training a sequence-to-sequence model and within introduction of applications of softmax smoothing to an N-best generation of the MBR training of the sequence-to-sequence model.

Abstract: Disclosed herein are compounds, compositions, and methods for decreasing TIGIT mRNA and protein expression. Such methods are useful to treat, prevent, or ameliorate TIGIT associated diseases, disorders, and conditions.

Abstract: Methods and apparatuses are provided for performing sequence to sequence (Seq2Seq) speech recognition training performed by at least one processor. The method includes acquiring a training set comprising a plurality of pairs of input data and target data corresponding to the input data, encoding the input data into a sequence of hidden states, performing a connectionist temporal classification (CTC) model training based on the sequence of hidden states, performing an attention model training based on the sequence of hidden states, and decoding the sequence of hidden states to generate target labels by independently performing the CTC model training and the attention model training.

Abstract: An intelligent web-enabled plant growing system controls environmental conditions inside it to grow sprouts, microgreens and plants from seeds placed in it without much intervention of humans through the growing process. Photographs of the sprouts, microgreens and plants are taken periodically by a camera. A computer then uploads the photographs to a server. The server runs artificial intelligence programs to determine the state of the growth and suggests back to the computer how to operate the controls for environment. A user application program on a remote computer or mobile device reports the state of the box, enables the user to buy seeds, enables the user to make own recipes, and enables multiple users to share their growing experience and custom recipes.