Abstract: Techniques are provided for pre-processing enhancement of a speech signal. A methodology implementing the techniques according to an embodiment includes performing de-reverberation processing on signals received from an array of microphones, the signals comprising speech and noise. The method also includes generating time-frequency masks (TFMs) for each of the signals. The TFMs indicate the probability that a time-frequency component of the signal associated with that TFM element includes speech. The TFM generation is based on application of a recurrent neural network to the signals. The method further includes generating steering vectors based on speech covariance matrices and noise covariance matrices. The TFMs are employed to filter speech components of the signals, for calculation of the speech covariance, and noise components of the signals for calculation of the noise covariance.

Abstract: A mechanism, method, and computer readable medium to enhance speech enabled devices. The method comprising receiving, by an ultrasonic attack detector of a speech enabled device, an audio stream from one or more microphones and a segmentation signal from a keyword detector indicating a location of a detected keyword within the audio stream, preprocessing, by the ultrasonic attack detector, a segmented portion of the audio stream including the detected keyword to obtain a spectrogram, and executing, by the ultrasonic attack detector, a neural network classifier using the spectrogram as input, the neural network classifier to discern real human speech from intermodulation distortion products resulting from ultrasonic attacks on the speech enabled device.

Abstract: Techniques related to keyphrase detection for applications such as wake on voice are disclosed herein. Such techniques may have high accuracy by using scores of phone positions in triphones to select which triphones to use with a rejection model, using context-related phones for the rejection model, adding silence before keyphrase sounds for a keyphrase model, or any combination of these.

Abstract: Techniques related to a method and system of audio false keyphrase rejection using speaker recognition are described herein. Such techniques use speaker recognition of a computer originated voice to omit actions triggered when a keyphrase is present in captured audio and omitted when speech of the captured audio was spoken by the computer originated voice.

Abstract: Techniques are provided for efficient acoustic event detection with reduced resource consumption. A methodology implementing the techniques according to an embodiment includes calculating frames of power spectra based on segments of received acoustic signals. The method further includes two processes, one for detecting impulsive acoustic events and another for detecting continuous acoustic events. The first process includes generating impulsive acoustic event features associated with first and second power spectrum frames, applying a neural network classifier to the impulsive acoustic event features to generate event scores, and detecting an impulsive acoustic event based on those event scores.

Abstract: Techniques are provided for acoustic event detection. A methodology implementing the techniques according to an embodiment includes extracting acoustic features from a received audio signal. The acoustic features may include, for example, one or more short-term Fourier transform frames, or other spectral energy characteristics, of the audio signal. The method also includes applying a trained classifier to the extracted acoustic features to identify and label acoustic event subparts of the audio signal and to generate scores associated with the subparts. The method further includes performing sequence decoding of the acoustic event subparts and associated scores to detect target acoustic events of interest based on the scores and temporal ordering sequence of the event subparts. The classifier is trained on acoustic event subparts that are generated through unsupervised subspace clustering techniques applied to training data that includes target acoustic events.

Abstract: Techniques are provided for pre-processing enhancement of a speech signal. A methodology implementing the techniques according to an embodiment includes performing de-reverberation processing on signals received from an array of microphones, the signals comprising speech and noise. The method also includes generating time-frequency masks (TFMs) for each of the signals. The TFMs indicate the probability that a time-frequency component of the signal associated with that TFM element includes speech. The TFM generation is based on application of a recurrent neural network to the signals. The method further includes generating steering vectors based on speech covariance matrices and noise covariance matrices. The TFMs are employed to filter speech components of the signals, for calculation of the speech covariance, and noise components of the signals for calculation of the noise covariance.

Abstract: The present invention relates to a method for identifying material types of spatial objects characterized in that the method comprising obtaining an acoustic signal from each identified object by deforming the objects mechanically, recording said acoustic signal and comparing it to an acoustic model being obtained on the basis of analysis of reference objects of multiple material types. The present invention also relates to a device for identifying material types of spatial objects, comprising a deformation chamber (K), a mechanical deformation system (F), at least one electro-acoustic transducer (1), an acoustic signal recording assembly (2) and a data processing unit (3) with installed acoustic model being obtained on the basis of analysis of reference objects of multiple material types.

Abstract: The present invention relates to a method for identifying material types of spatial objects characterized in that the method comprising obtaining an acoustic signal from each identified object by deforming the objects mechanically, recording said acoustic signal and comparing it to an acoustic model being obtained on the basis of analysis of reference objects of multiple material types. The present invention also relates to a device for identifying material types of spatial objects, comprising a deformation chamber (K), a mechanical deformation system (F), at least one electro-acoustic transducer (1), an acoustic signal recording assembly (2) and a data processing unit (3) with installed acoustic model, being obtained on the basis of analysis of reference objects of multiple material types.