Abstract: Some embodiments are directed to a federated learning system. A federated model is trained on respective local training datasets of respective multiple edge devices. In an iteration, an edge device obtains a current federated model, determines a model update for the current federated model based on the local training dataset, and sends out the model update. The edge device determines the model update by applying the current federated model to a training input to obtain at least a model output for the training input; if the model output does not match a training output corresponding to the training input, include the training input in a subset of filtered training inputs to be used in the iteration; and determining the model update by training the current federated model on only the subset of filtered training inputs.

Abstract: Some embodiments are directed to training a model, e.g., a medical model. The training uses multiple model updates received from multiple client systems. At least some of the multiple client train on training sets that indicate values for different features. The model updates are aggregated in an aggregated model, for which feature weights are obtained. The feature weights provide information on the relative importance of the multiple features for the aggregated model's output.

Abstract: According to an aspect, there is provided a computer implemented method in a central sewer of selecting a training dataset with which to train a model using a distributed machine learning process, wherein the training dataset is to comprise medical data that satisfies one or more clinical requirements and wherein training data in the training dataset is located at a plurality of clinical sites. The method comprises requesting (302) from each of the clinical sites, metadata describing features of matching data at the respective clinical site that satisfies the one or more clinical requirements. The method then comprises determining (304), from the metadata, a measure of variation of the features of the matching data. Based on the measure of variation, the method then comprises selecting (306) training data for the training dataset from the matching data using the metadata.

Abstract: According to an aspect, there is provided a computer-implemented method of structuring content for training an artificial intelligence model, the method comprising: receiving (S11) input content associated with medical device documentation; converting (S12) the input content to a data interchange format; extracting (S13) a plurality of key terms from the converted input content; extracting (S14) a plurality of key phrases from the converted input content; receiving (S15) validation of the key terms and the key phrases from a supervisor; and building (S16) a dialogue, for training the artificial intelligence model, based on at least some of the validated key terms and the validated key phrases, wherein the dialogue comprises a series of statements.

Abstract: The present disclosure relates to preserving context in a conversation between a user (101) and a digital assistant device (102). During training, the digital assistant device (102) is provided with a plurality of conversations having a plurality of dialogues. Each of the plurality of dialogue is assigned an ID based on a context. Further, two or more test queries having a same context is provided as input and the two are more queries are assigned an ID based on the context. Thereafter, the digital assistant device (102) is configured to retrieve one or more dialogues from the plurality of dialogues where the ID of the one or more dialogues match the ID of the two or more queries. In real-time, one or more queries are received and based on a context of the one or more queries, one or more dialogues are retrieved and are provided to the user.

Abstract: The invention discloses an apparatus for stroke characterization. The apparatus comprises a processor. The processor is configured to receive image data representing a three-dimensional image of a head of a subject; identify a region within the image data corresponding to bone in the head of the subject; apply an adjustment to the image data to compensate for effects resulting from radiation scattered from the bone during acquisition of the image data; generate a two-dimensional image based on the adjusted image data; and provide the generated two-dimensional image as an input to a predictive model to identify a region of interest for stroke characterization in the two-dimensional image.

Abstract: A system and method that enhances spoken utterances and provides entertainment by capturing one or more microphone signals containing echo and decomposing the one or more microphone signals into a plurality of signal paths through a synthesizer that adds or makes non-linear modifications to some of the captured one or more microphone signals. The system and method and estimates multiple echo paths from each of the one the one or more microphones. The system and method processes the captured microphone signals in response to the estimated plurality of echo paths by subtracting the echo contributions of each of the plurality of echo paths from the captured one or more microphone signals. The system and method also provide signal separation and post processing functions that renders speech recognition gaming applications.

Abstract: A method for generating a textual description from a medical image, comprising: receiving a medical image having a first modality to a system configured to generate a textual description of the medical image; determining, using an imaging modality classification module, that the first modality is a specific one of a plurality of different modalities; determining, using an anatomy classification module, that the medical image comprises information about a specific portion of an anatomy; identifying, by an orchestrator module based at least on the determined first modality, which of a plurality of different text generation models to utilize to generate a textual description from the medical image; generating, by a text generation module utilizing the identified text generation model, a textual description from the medical image; and reporting, via a user interface of the system, the generated textual description.

Abstract: A digital audio system, which may be an integrated audio system of a vehicle, includes a microphone, a loudspeaker, and a processor. The processor is operable to modify digital audio signals based on tuning parameters to compensate for acoustic characteristics of the environment. The processor is further operable to detect that modification of digital audio signals based on the tuning parameters cannot fully compensate for the acoustic characteristics of the environment. The detection is based on an audio signal corresponding to audio captured from the environment and/or on the values of the tuning parameters. Upon detecting that modification of digital audio signals based on the tuning parameters cannot fully compensate for the acoustic characteristics of the environment, an indication is provided that the audio system should be calibrated to adjust the tuning parameters to the environment. Related methods and computer-readable media are also disclosed.

Abstract: A digital audio system, which may be an integrated audio system of a vehicle, includes a microphone, a loudspeaker, and a processor. The processor is operable to modify digital audio signals based on tuning parameters to compensate for acoustic characteristics of the environment. The processor is further operable to detect that modification of digital audio signals based on the tuning parameters cannot fully compensate for the acoustic characteristics of the environment. The detection is based on an audio signal corresponding to audio captured from the environment and/or on the values of the tuning parameters. Upon detecting that modification of digital audio signals based on the tuning parameters cannot fully compensate for the acoustic characteristics of the environment, an indication is provided that the audio system should be calibrated to adjust the tuning parameters to the environment. Related methods and computer-readable media are also disclosed.

Abstract: A multimodal system and process processed one or more sensor signals and extracts features from the one or more sensor signals through a spatiotemporal correlation between consecutive frames of an image or video sequence. The multimodal system and process determines the movement and direction of the features through an image subtraction, or a coherence measure and synthesizes a musical instrument signal in response to the detected movement and direction or the triggers. The imaginary musical instrument signal is added to an infotainment signal within a vehicle.

Abstract: A method for generating a textual description from a medical image, comprising: receiving a medical image having a first modality to a system configured to generate a textual description of the medical image; determining, using an imaging modality classification module, that the first modality is a specific one of a plurality of different modalities; determining, using an anatomy classification module, that the medical image comprises information about a specific portion of an anatomy; identifying, by an orchestrator module based at least on the determined first modality, which of a plurality of different text generation models to utilize to generate a textual description from the medical image; generating, by a text generation module utilizing the identified text generation model, a textual description from the medical image; and reporting, via a user interface of the system, the generated textual description.

Abstract: A multimodal system and process processed one or more sensor signals and extracts features from the one or more sensor signals through a spatiotemporal correlation between consecutive frames of an image or video sequence. The multimodal system and process determines the movement and direction of the features through an image subtraction, or a coherence measure and synthesizes a musical instrument signal in response to the detected movement and direction or the triggers. The imaginary musical instrument signal is added to an infotainment signal within a vehicle.

Abstract: A system and method that enhances spoken utterances and provides entertainment by capturing one or more microphone signals containing echo and decomposing the one or more microphone signals into a plurality of signal paths through a synthesizer that adds or makes non-linear modifications to some of the captured one or more microphone signals. The system and method and estimates multiple echo paths from each of the one the one or more microphones. The system and method processes the captured microphone signals in response to the estimated plurality of echo paths by subtracting the echo contributions of each of the plurality of echo paths from the captured one or more microphone signals. The system and method also provide signal separation and post processing functions that renders speech recognition gaming applications.

Abstract: A system and method that enhances spoken utterances and provides entertainment by capturing one or more microphone signals containing echo and decomposing the one or more microphone signals into a plurality of signal paths through a synthesizer that adds or makes non-linear modifications to some of the captured one or more microphone signals. The system and method and estimates multiple echo paths from each of the one the one or more microphones. The system and method processes the captured microphone signals in response to the estimated plurality of echo paths by subtracting the echo contributions of each of the plurality of echo paths from the captured one or more microphone signals. The system and method also provide signal separation and post processing functions that renders speech recognition gaming applications.

Abstract: A system and method for mitigating audio feedback may calculate a smoothed frequency spectrum of an audio signal. Previously detected candidate feedback tones may be obtained. Candidate feedback tones may be determined responsive to a frequency spectrum of the audio signal, the smoothed frequency spectrum and the previously detected candidate feedback tones. One or more signal characteristics associated with the audio signal and feedback coefficients associated with the candidate feedback tones may be obtained. The feedback coefficients may be modified responsive to the one or more signal characteristics. Actionable feedback tones may be determined responsive to the associated feedback coefficients exceeding a respective feedback threshold. Feedback suppression coefficients associated with each of the determined actionable feedback tones may be generated and may be utilized to suppress the actionable feedback tones.

Abstract: A system and method for mitigating audio feedback may calculate a smoothed frequency spectrum of an audio signal. Previously detected candidate feedback tones may be obtained. Candidate feedback tones may be determined responsive to a frequency spectrum of the audio signal, the smoothed frequency spectrum and the previously detected candidate feedback tones. One or more signal characteristics associated with the audio signal and feedback coefficients associated with the candidate feedback tones may be obtained. The feedback coefficients may be modified responsive to the one or more signal characteristics. Actionable feedback tones may be determined responsive to the associated feedback coefficients exceeding a respective feedback threshold. Feedback suppression coefficients associated with each of the determined actionable feedback tones may be generated and may be utilized to suppress the actionable feedback tones.

Abstract: A reverberation device is provided for adding a reverberation signal to an audio input signal. The device consists of a late reflection generator and an early reflection generator which share common delay line memory. Sparsely selected taps are utilized to generate an early reflection signal from the input signal as it passes through the delay line. Subsequently, the delayed signal is attenuated and recirculated, and multiple randomly selected taps are utilized to generate a late reflection signal with enhanced echo density. The late reflection and early reflection signals are combined with the input signal to generate an output signal with added reverberation.