Abstract: A method for training a speech recognition model includes obtaining a multilingual text-to-speech (TTS) model. The method also includes generating a native synthesized speech representation for an input text sequence in a first language that is conditioned on speaker characteristics of a native speaker of the first language. The method also includes generating a cross-lingual synthesized speech representation for the input text sequence in the first language that is conditioned on speaker characteristics of a native speaker of a different second language. The method also includes generating a first speech recognition result for the native synthesized speech representation and a second speech recognition result for the cross-lingual synthesized speech representation. The method also includes determining a consistent loss term based on the first speech recognition result and the second speech recognition result and updating parameters of the speech recognition model based on the consistent loss term.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer-readable media, for speech recognition using multi-dialect and multilingual models. In some implementations, audio data indicating audio characteristics of an utterance is received. Input features determined based on the audio data are provided to a speech recognition model that has been trained to output score indicating the likelihood of linguistic units for each of multiple different language or dialects. The speech recognition model can be one that has been trained using cluster adaptive training. Output that the speech recognition model generated in response to receiving the input features determined based on the audio data is received. A transcription of the utterance generated based on the output of the speech recognition model is provided.

Abstract: A method in an illustrative embodiment includes determining that a first object authenticated by an electronic device is accessing the electronic device. The method further includes, in response to a second object being detected within a detection range using a radar of the electronic device, determining, based on a detected signal, that the second object is a person. The method further includes determining a distance and an angle between the second object and the electronic device based on an azimuth signal in the detected signal. The method further includes in response to determining that the distance is less than a distance threshold and the angle is less than an angle threshold, determining, based on the biological feature signal, whether the second object is trustworthy. The method further includes deauthenticating the first object in response to determining that the second object is untrustworthy.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for using a virtual desktop. A method in one embodiment includes receiving, at a first edge node in a plurality of edge nodes, an instruction from a first set of input devices in a plurality of peripheral devices. The instruction is for use of a first virtual desktop deployed on the first edge node. The method further includes: using the first virtual desktop based on the instruction by using resources at the first edge node. The method further includes: sending data to an output device in the plurality of peripheral devices, wherein the data is associated with the use of the first virtual desktop. The solution for using a virtual desktop of the present application enables the use of a virtual desktop using resources at an edge node without requiring a client.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for video processing. The method includes: converting, based on a sample frame in a first video having a first resolution as well as a template, a first group of image frames in the first video that correspond to the sample frame to a second group of image frames that are more similar to the template. The method also includes: converting the second group of image frames to a third group of image frames having a higher resolution to generate a higher-resolution video. In this manner, low-resolution image frames can be first converted to image frames that are more suitable for resolution conversion, so that a high-resolution video of a higher quality can be obtained when reconstructing the high-resolution video.

Abstract: A method for training a speech recognition model includes receiving a set of training utterance pairs each including a non-synthetic speech representation and a synthetic speech representation of a same corresponding utterance. At each of a plurality of output steps for each training utterance pair in the set of training utterance pairs, the method also includes determining a consistent loss term for the corresponding training utterance pair based on a first probability distribution over possible non-synthetic speech recognition hypotheses generated for the corresponding non-synthetic speech representation and a second probability distribution over possible synthetic speech recognition hypotheses generated for the corresponding synthetic speech representation. The first and second probability distributions are generated for output by the speech recognition model.

Abstract: Embodiments of the disclosure include a method, a device, and a computer program product for video processing. This method includes: selecting frames having features of a first type from a first instance of a video as a first candidate set, the first instance having a first resolution; generating a set of training frames based at least on the first candidate set; acquiring a set of corresponding frames for the set of training frames in a second instance of the video, the second instance having a second resolution lower than the first resolution; and determining, using the set of training frames and the set of corresponding frames, a conversion parameter for conversion from the second resolution to a third resolution. This solution provides a smaller-scale and higher-quality training set for the training of a video conversion model, thus improving the quality of training while saving computational resources and increasing training speed.

Abstract: Systems, methods, and non-transitory computer-readable media can determine that a user has requested a service that involves payment. Information associated with the user can be analyzed. It can be determined, based on the information associated with the user, that the user is eligible for a subsequent payment option. The subsequent payment option can be provided to the user as a line of credit applicable toward payment for the service.