SYSTEM AND METHOD FOR ESTIMATING HORMONE LEVEL AND PHYSIOLOGICAL CONDITIONS BY ANALYSING SPEECH SAMPLES

• US 20180322893A1
• Filed: 05/01/2018
• Published: 11/08/2018
• Est. Priority Date: 05/03/2017
• Status: Active Grant

• Current Assignee: Ajit Arun Zadgaonkar, Arun Shrihari Zadgaonkar
• Original Assignee: Ajit Arun Zadgaonkar, Arun Shrihari Zadgaonkar
• Inventors: Zadgaonkar, Ajit Arun, Zadgaonkar, Arun Shrihari
• Granted Patent: US 10,593,351 B2
• Time in Patent Office: Days
• CPC Class Codes: A61B 5/14546 for measuring analytes not otherwise provided for, e.g. ions, cytochromes
• CPC Class Codes: A61B 5/4227 endocrine glands, i.e. thyroid, adrenals, hypothalamic, pituitary
• CPC Class Codes: A61B 5/4803 Speech analysis specially adapted for diagnostic purposes
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 17/26 Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
• CPC Class Codes: G10L 2015/025 Phonemes, fenemes or fenones being the recognition units
• CPC Class Codes: G10L 25/15 the extracted parameters being formant information
• CPC Class Codes: G10L 25/66 for extracting parameters related to health condition detecting or measuring for diagnostic purposes A61B5/00

Abstract

The present disclosure describes a system and method for estimating hormone levels and physiological conditions of a user by analysing speech samples of said user. A user device of the user may record specifics of speech and use these specifics of speech as a speech sample of user'"'"'s utterance. The user device may transmit the speech samples to a backend system. The system may isolate phonation segments from the speech samples. The system may filter the one or more phonation segments. The system may isolate uttered speech segments from the one or more phonation segments. The system may perform an acoustic-phonetic analysis of the uttered speech segments. The acoustic-phonetic analysis may use plurality of features for the analysis. The IPA phonemes may be used to derive speech markers that correspond to specific hormones and levels thereof. The system may generate a hormone level report which is transmitted to the user.

Claims

1. A system (101) for estimating hormone level and physiological conditions of a user, the system comprising:

a processor (201); and

a memory (203) coupled with the processor, wherein the processor is configured to execute a plurality of programmed instructions stored in the memory (203), the plurality of programmed instructions comprising instructions for;

interactively guiding a user to utter specifics of a speech via a user device (104) communicatively coupled with the processor (201);

recording, via the user device, one or more speech samples associated with the user based upon the specifics of speech uttered by the user;

isolating one or more phonation segments from the one or more speech samples;

filtering one or more phonation segments to remove noise from the one or more phonation segments;

isolating one or more uttered speech segments from the one or more phonation segments filtered;

performing acoustic-phonetic analysis of the one or more uttered speech segments to extract one or more speech features; and

determining one or more speech markers and the corresponding hormone level of the user based upon the one or more speech features.

2. The system according to claim 1, wherein the specifics of speech comprises attributes selected from a group comprising of a frequency and formants, and wherein the specifics of speech comprises one or more of words, letters, and phonations.

3. The system according to claim 1, wherein the one or more speech features comprises utterance of the user, vowel and consonants under isolated and/or vowel consonant (VC), consonant Vowel (CV) environment, formants of the utterance, pitch of the utterance, vocal intensity of the utterance, speech quality of the utterance, vowel onset point (VOP), energy transition, values of dispersion and bandwidth, voice onset time (VOT), articulatory rate, shimmer, jitter, and spectrogram.

4. The system according to claim 1, wherein the speech markers corresponding to the hormone levels comprises Human chorionic gonadotropin (hcG) speech marker, estrogen speech marker, progesterone speech marker, Luteinizing hormone (LH) speech marker, and Follicle Stimulating Hormone (FSH) speech marker, and wherein the speech markers and the corresponding hormone levels are determined using International Phonetic alphabets (IPA) phonemes selected from a group comprising semi-vowel, bilabial nasal consonant and voiced bilabial consonant.

5. The system according to claim 4, wherein a hormone level corresponding to the Follicle Stimulating Hormone (FSH) is an index of a speech marker consisting of a ratio of the pitch and formants of the semi-vowels, bilabial nasal consonant and voiced bilabial consonants.

6. The system according to claim 4, wherein a hormone level corresponding to the Luteinizing hormone (LH) is an index of a speech marker consisting of a ratio of formants of the semi-vowels, bilabial nasal consonant and voiced bilabial consonants.

7. The system according to claim 4, wherein a hormone level corresponding to the estrogen hormone is an index of a speech marker consisting of half of the difference of the formant values multiplied by frequency perturbation factor of the semi-vowels, bilabial nasal consonants and voiced bilabial consonants.

8. The system according to claim 4, wherein a hormone level corresponding to the progesterone hormone is determined from the speech markers of the semi-vowels, from the ratio of formants.

9. The system according to claim 4, wherein a hormone level corresponding to the Human chorionic gonadotropin (hCG) hormone is an index of a speech marker consisting of sum of pitch and formant values of the semi-vowels, bilabial nasal consonant and voiced bilabial consonants.

10. The system according to claim 1, wherein the programmed instructions further comprises programmed instructions for generating reports based upon the estimation of hormonal levels of the user, wherein the reports generated comprises separate reports for one or more of estrogen, progesterone, Luteinizing Hormone (LH), Follicle Stimulating Hormone (FSH) and other human body hormones.

11. A method for estimating hormone level and physiological conditions of a user, the method comprising:

interactively guiding, via a user device communicatively coupled with a processor, a user to utter specifics of a speech;

recording, via the user device, one or more speech samples associated with the user based upon the specifics of speech uttered by the user;

isolating, via the processor, one or more phonation segments from the one or more speech samples;

filtering, via the processor, one or more phonation segments to remove noise from the one or more phonation segments;

isolating, via the processor, one or more uttered speech segments from the one or more phonation segments filtered;

performing, via the processor, acoustic-phonetic analysis of the one or more uttered speech segments to extract one or more speech features; and

determining, via the processor, one or more speech markers and the corresponding hormone level of the user based upon the one or more speech features.

12. The method according to claim 11, further comprising generating, via the processor, reports based upon the estimation of hormonal levels of the user, wherein the reports generated comprises separate reports for one or more of estrogen, progesterone, Luteinizing Hormone (LH), Follicle Stimulating Hormone (FSH) and other human body hormones.