Methods and apparatuses for detecting user speech are described. In one example, a method for detecting user speech includes receiving a microphone output signal corresponding to sound received at a microphone and identifying a spoken vowel sound in the microphone signal. The method further includes outputting an indication of user speech detection responsive to identifying the spoken vowel sound.

A multimedia information processing method includes: parsing multimedia information to separate an audio from the multimedia information; converting the audio to obtain a mel spectrogram corresponding to the audio; determining, according to the mel spectrogram corresponding to the audio, an audio feature vector corresponding to the audio; and determining, based on an audio feature vector corresponding to a source audio in source multimedia information and an audio feature vector corresponding to a target audio in target multimedia information, a similarity between the target multimedia information and the source multimedia information.

A multimedia information processing method includes: parsing multimedia information to separate an audio from the multimedia information; converting the audio to obtain a mel spectrogram corresponding to the audio; determining, according to the mel spectrogram corresponding to the audio, an audio feature vector corresponding to the audio; and determining, based on an audio feature vector corresponding to a source audio in source multimedia information and an audio feature vector corresponding to a target audio in target multimedia information, a similarity between the target multimedia information and the source multimedia information.

Method and apparatus obtain information about a patient and/or a respiratory therapy system that is configured to deliver respiratory therapy to the patient. The respiratory therapy system may include a flow generator configured to generate a supply of pressurized air along an air circuit to a patient interface. A sound signal representing a sound in the air circuit may be processed to obtain cepstrum data. A time series of delay estimates based on acoustic signatures of the cepstrum data may be generated. Each acoustic signature may represent a reflection of sound from a patient interface along the air circuit. Variation in the time series of delay estimates may be analysed. One or more output indicators based on the variation may be generated. The one or more output indicators may concern patient and/or system status.

Method and apparatus obtain information about a patient and/or a respiratory therapy system that is configured to deliver respiratory therapy to the patient. The respiratory therapy system may include a flow generator configured to generate a supply of pressurized air along an air circuit to a patient interface. A sound signal representing a sound in the air circuit may be processed to obtain cepstrum data. A time series of delay estimates based on acoustic signatures of the cepstrum data may be generated. Each acoustic signature may represent a reflection of sound from a patient interface along the air circuit. Variation in the time series of delay estimates may be analysed. One or more output indicators based on the variation may be generated. The one or more output indicators may concern patient and/or system status.

A method to detect and analyze cough parameters remotely and in the background through the user's device in everyday life. The method involves consecutive or selective execution of 4 stages, which solve a variety of the following tasks: Detection of coughing events in sound, in environmental noise. Separation of the sound containing the cough into the sound of the cough and the rest of the sounds, even when extraneous sounds occurred during the cough. Identify the user by the sound of the cough to prevent analyzing the sounds of the cough that do not belong to the user (for example when another person coughs nearby). Assessment of cough characteristics (wet/dry, severity, duration, number of spasms, etc.). The method can work on wearable devices, smart devices, personal computers, laptops in 24/7 mode or in a selected range of time, and has a high energy efficiency.

A method to detect and analyze cough parameters remotely and in the background through the user's device in everyday life. The method involves consecutive or selective execution of 4 stages, which solve a variety of the following tasks: Detection of coughing events in sound, in environmental noise. Separation of the sound containing the cough into the sound of the cough and the rest of the sounds, even when extraneous sounds occurred during the cough. Identify the user by the sound of the cough to prevent analyzing the sounds of the cough that do not belong to the user (for example when another person coughs nearby). Assessment of cough characteristics (wet/dry, severity, duration, number of spasms, etc.). The method can work on wearable devices, smart devices, personal computers, laptops in 24/7 mode or in a selected range of time, and has a high energy efficiency.

A sound source separation program causes a computer to acquire an acoustic signal, convert the acquired acoustic signal from a time region to a frequency region, and perform sound source separation on the acoustic signal converted to the frequency region by performing updating based on elementary row operation on a demixing matrix to iteratively minimize an objective function including a quadratic form of a separation vector and a determinant of the demixing matrix.

The present disclosure provides a method and a device of denoising a voice signal. The method portion includes the following steps: filtering out an environmental noise signal in an original input signal according to an interference signal related to the environmental noise signal in the original input signal to obtain a first voice signal; obtaining a sample signal matching the first voice signal from a voice signal sample library; and filtering out other noise signal in the first voice signal according to the sample signal matching the first voice signal, to obtain an effective voice signal. The method provided by the present disclosure may effectively filter out the environmental noise signal and other noise signal in the voice signal.

The present disclosure provides a method and a device of denoising a voice signal. The method portion includes the following steps: filtering out an environmental noise signal in an original input signal according to an interference signal related to the environmental noise signal in the original input signal to obtain a first voice signal; obtaining a sample signal matching the first voice signal from a voice signal sample library; and filtering out other noise signal in the first voice signal according to the sample signal matching the first voice signal, to obtain an effective voice signal. The method provided by the present disclosure may effectively filter out the environmental noise signal and other noise signal in the voice signal.