Abstract: Embodiments of the present invention provide methods and systems for performing automatic diarisation of sound recordings including speech from one more speakers. The automatic diarisation has a development or training phase and a utilisation or evaluation phase. In the development or training phase background models and hyperparameters are generated from already annotated sound recordings. These models and hyperparameters are applied during the evaluation or utilisation phase to diarise new or not previously diarised or annotated recordings.

Abstract: Embodiments of the present invention provide methods and systems for performing automatic diarisation of sound recordings including speech from one more speakers. The automatic diarisation has a development or training phase and a utilisation or evaluation phase. In the development or training phase background models and hyperparameters are generated from already annotated sound recordings. These models and hyperparameters are applied during the evaluation or utilisation phase to diarise new or not previously diarised or annotated recordings.

Abstract: Embodiments of the present invention provide methods and systems for performing automatic diarisation of sound recordings including speech from one more speakers. The automatic diarisation has a development or training phase and a utilisation or evaluation phase. In the development or training phase background models and hyperparameters are generated from already annotated sound recordings. These models and hyperparameters are applied during the evaluation or utilisation phase to diarise new or not previously diarised or annotated recordings.

Abstract: Embodiments of the present invention provide methods and systems for performing automatic diarisation of sound recordings including speech from one more speakers. The automatic diarisation has a development or training phase and a utilisation or evaluation phase. In the development or training phase background models and hyperparameters are generated from already annotated sound recordings. These models and hyperparameters are applied during the evaluation or utilisation phase to diarise new or not previously diarised or annotated recordings.

Abstract: A parameter quantifying deviation from Gaussianity distribution of a patient's sounds, in the form of a non-Gaussianity distribution index, may be used to assist in diagnosis of sleep dysfunction such as OSAHS. A method for diagnosing a sleeping disorder of a subject includes processing a digitized audio signal from the subject with at least one electronic processor. The processing includes estimating a parameter quantifying deviation from Gaussianity distribution of the audio signal. in the form of a non-Gaussianity Index (NGI). The NGI value is then applied to a diagnostic model. The presence of a sleeping disorder is then indicated based on the output of the diagnostic model.

Abstract: A parameter quantifying deviation from Gaussianity distribution of a patient's sounds, in the form of a non-Gaussianity distribution index, may be used to assist in diagnosis of sleep dysfunction such as OSAHS. A method for diagnosing a sleeping disorder of a subject includes processing a digitized audio signal from the subject with at least one electronic processor. The processing includes estimating a parameter quantifying deviation from Gaussianity distribution of the audio signal. in the form of a non-Gaussianity Index (NGI). The NGI value is then applied to a diagnostic model. The presence of a sleeping disorder is then indicated based on the output of the diagnostic model.

Abstract: An apparatus for diagnosing sleep disorders such as OASHS from snore sounds includes a segmentation module (126) coupled to a data logger (124) to provide segments of the digitized audio signal to a Snore Segment Identifier (128). A total airways response (TAR) module (130), pitch calculator (132), and MFCC calculator (134) are each coupled to an output side of the snore segment identifier module (128). Each of these modules is respectively arranged to calculate pitch, bispectrum, diagonal slice, and MFCC parameters for the snore segments received from the snore segment identifier (128). Similarly, the NGI calculator (136) produces a non-Gaussianity index for the digitized audio signal. A classification module (144) is arranged to process the calculated parameters and compare a resulting diagnosis probability to a predetermined threshold value.

Abstract: An apparatus for diagnosing sleep disorders such as OASHS from snore sounds includes a segmentation module (126) coupled to a data logger (124) to provide segments of the digitized audio signal to a Snore Segment Identifier (128). A total airways response (TAR) module (130), pitch calculator (132) and MFCC calculator (134) are each coupled to an output side of the snore segment identifier module (128). Each of these modules is respectively arranged to calculate pitch, bispectrum, diagonal slice and MFCC parameters for the snore segments received from the snore segment identifier (128). Similarly, the NGI calculator (136) produces a non-Gaussianity index for the digitized audio signal. A classification module (144) is arranged to process the calculated parameters and compare a resulting diagnosis probability to a predetermined threshold value.