Health-related information generation and storage
A detected sound signal may comprise speech or non-verbal sounds, and many non-verbal sounds contain health information. If the speech, or a non-verbal sound containing health information, was produced by an enrolled user, data relating to the sound can be stored in a storage element. A system also comprises a data modification block, for obfuscating received data to provide an obfuscated version of the stored data. The system then has a first access mechanism, for controlling access to the stored data such that only an authorised user can obtain access to said stored data, and a second access mechanism, for controlling access to said stored data such that the second access mechanism only provides access to the obfuscated version of the stored data.
1. A method of monitoring an enrolled user's health, comprising:
receiving an audio signal representing speech;
performing a speaker recognition process on the received audio signal to identify a speaker;
if it is determined that the identified speaker is an enrolled user, obtaining health marker features from the speech represented by the audio signal; and
storing the obtained health marker features together with data indicating the identified speaker in a secure storage element.
2. A method according to claim 1 , comprising extracting features of the received audio signal, and obtaining said health marker features from the extracted features.
3. A method according to claim 2 , comprising performing said speaker recognition process using the same extracted features as are used to obtain said health markers.
4. A method according to claim 1 , comprising applying a time stamp to each health marker feature, and storing the time stamp with the health marker feature in the secure storage element.
5. A method according to claim 1 , comprising obtaining the health marker features over at least one time period, and storing data identifying the time period with the health marker feature in the secure storage element.
6. A method according to claim 1 , wherein the health marker features comprise one or more of: a fundamental frequency of the speech; Jitter; Shimmer; a Long Term Average Spectrum of the speech; an articulation rate; a maximal phonation time; a Noise-to-Harmonic Ratio (NHR); a Harmonic-to-Noise Ratio (HNR); and a Cepstral peak prominence.
7. A system comprising:
an input for receiving an audio signal representing speech; and
a processor, for performing a speaker recognition process on the received audio signal to identify a speaker; if it is determined that the identified speaker is an enrolled user, obtaining health marker features from the speech represented by the audio signal; and storing the obtained health marker features together with data indicating the identified speaker in a secure storage element.
8. A device comprising a system according to claim 7 , wherein the device comprises: a mobile computing device, a games console, a remote control device, a home automation controller or a domestic appliance, a toy, a machine, an audio player, a video player, or a mobile telephone.
9. A computer program product, comprising computer-readable code for causing a programmed processor to perform the method of claim 1 .
10. A computer program product according to claim 9 , comprising a tangible computer-readable medium for storing said code.