Abstract: Implementations are directed to efficient federated learning of machine learning (ML) model(s) through on-the-fly decompression and compression of model parameters, of the ML model(s), when facilitating forward propagation and/or back propagation at client device(s). For example, implementations can transmit, from a remote system to a client device, a compressed on-device ML model that includes some compressed parameters. Further, the client device can, in performing forward propagation and/or back propagation using the on-device ML model, decompress those compressed parameters on-the-fly as the parameters are needed for the propagation. The propagation will utilize the decompressed parameters that were decompressed on the fly.

Abstract: A method includes receiving distillation data including a plurality of out-of-domain training utterances. For each particular out-of-domain training utterance of the distillation data, the method includes generating a corresponding augmented out-of-domain training utterance, and generating, using a teacher ASR model trained on training data corresponding to a target domain, a pseudo-label corresponding to the corresponding augmented out-of-domain training utterance. The method also includes distilling a student ASR model from the teacher ASR model by training the student ASR model using the corresponding augmented out-of-domain training utterances paired with the corresponding pseudo-labels generated by the teacher ASR model.

Abstract: Information can be distilled from a global automatic speech recognition (ASR) model to a client ASR model. Many implementations include using an RNN-T model as the ASR model, where the global ASR model includes a global encoder, a joint network, a prediction network, and where the client ASR model includes a client encoder, the joint network, and the prediction network. Various implementations include using principal component analysis (PCA) while training the global ASR model to learn a mean vector and a set of principal components corresponding to the global ASR model. Additional or alternative implementations include training the client ASR model to generate one or more predicted coefficients of the global ASR model.

Abstract: A method includes receiving distillation data including a plurality of out-of-domain training utterances. For each particular out-of-domain training utterance of the distillation data, the method includes generating a corresponding augmented out-of-domain training utterance, and generating, using a teacher ASR model trained on training data corresponding to a target domain, a pseudo-label corresponding to the corresponding augmented out-of-domain training utterance. The method also includes distilling a student ASR model from the teacher ASR model by training the student ASR model using the corresponding augmented out-of-domain training utterances paired with the corresponding pseudo-labels generated by the teacher ASR model.

Abstract: Disclosed is an event detecting method, which is applied to an electronic system with at least one event detecting module. The event detecting method comprises: (a) enabling at least one event detecting module; and (b) applying the event detecting module enabled in the step (a) to detect if any predetermined event occurs in a detecting range of the event detecting module.

Abstract: An electronic device controlling method and a user registration method are provided. In the electronic device controlling method, when a target device receives a first and a second control commands which are identical, but performed by different users simultaneously or separately, the target device performs a first predetermined operation based on an identity of the user performing the first control command, and performs a second predetermined operation based on an identity of the user performing the second control command. In the user registration method, a user registered identity model corresponding to a user to be registered is established according to identity information of the user, and is mapped to a user profile comprising a relationship between the control commands and the predetermined operations. By acquiring the registered information, the target device is able to perform the user dependent operations.

Abstract: An exemplary information sharing method is provided. The information sharing method includes a step of: broadcasting predetermined sharing information to a user terminal(s) within a preset region on a client side in response to a sharing request while an information sharing function is enable. An information sharing module, device and electronic apparatus using the same are also provided.