Technologies for robust crying detection using temporal characteristics of acoustic features
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G10L-025/51
G10L-025/24
G10L-025/72
G10L-025/27
출원번호
US-0979108
(2015-12-22)
등록번호
US-9899034
(2018-02-20)
발명자
/ 주소
Hofer, Joachim
Bocklet, Tobias
Stemmer, Georg
Pearce, David
Czyryba, Sebastian
Bauer, Josef G.
출원인 / 주소
Intel IP Corporation
대리인 / 주소
Barnes & Thornburg LLP
인용정보
피인용 횟수 :
0인용 특허 :
1
초록▼
Technologies for identifying sounds are disclosed. A sound identification device may capture sound data, and split the sound data into frames. The sound identification device may then determine an acoustic feature vector for each frame, and determine parameters based on how each acoustic feature var
Technologies for identifying sounds are disclosed. A sound identification device may capture sound data, and split the sound data into frames. The sound identification device may then determine an acoustic feature vector for each frame, and determine parameters based on how each acoustic feature varies over the duration of time corresponding to the frames. The sound identification device may then determine if the sound matches a pre-defined sound based on the parameters. In one embodiment, the sound identification device may be a baby monitor, and the pre-defined sound may be a baby crying.
대표청구항▼
1. A sound identification device for identifying sounds, the sound identification device comprising: a sound data capture module to acquire sound data;a sound frame determination module to determine a plurality of frames of sound data based on the sound data;a sound identification module to: determi
1. A sound identification device for identifying sounds, the sound identification device comprising: a sound data capture module to acquire sound data;a sound frame determination module to determine a plurality of frames of sound data based on the sound data;a sound identification module to: determine an acoustic feature matrix having two dimensions and comprising a plurality of first-dimension vectors and a plurality of second-dimension vectors, wherein each second-dimension vector of the plurality of second-dimension vectors corresponds to a corresponding frame of the plurality of frames and each first-dimension vector of the plurality of first-dimension vectors comprises an acoustic feature vector associated with the corresponding frame, and wherein each first-dimension vector of the plurality of first-dimension vectors is associated with a different acoustic feature;determine a plurality of temporal parameters for each first-dimension vector of the plurality of first-dimension vectors;determine, based on the pluralities of temporal parameters, whether the sound data corresponds to a pre-defined sound, wherein each of the plurality of temporal parameters is based on how the acoustic feature associated with the corresponding first-dimension vector changes over the time associated with the plurality of frames. 2. The sound identification device of claim 1, wherein each acoustic feature vector of the acoustic feature matrix comprises mel-frequency cepstrum coefficients. 3. The sound identification device of claim 1, wherein the pre-defined sound is a cry of an infant. 4. The sound identification device of claim 1, wherein the pre-defined sound is a cough. 5. The sound identification device of claim 4, further comprising a communication module to provide, based on the sound data corresponding to the cough, an alert to a user of the sound identification device and to provide, based on the sound data corresponding to the cough, a suggestion to the user. 6. The sound identification device of claim 1, wherein to determine the plurality of temporal parameters for each first-dimension vector of the plurality of first-dimension vectors comprises to determine the plurality of temporal parameters for each first-dimension vector of the plurality of first-dimension vectors by performing a Fourier-related transform on the corresponding first-dimension vector. 7. The sound identification device of claim 6, wherein the sound identification module is to select a subset of the plurality of temporal parameters, and wherein to determine, based on the plurality of temporal parameters, whether the sound data corresponds to the pre-defined sound comprises to determine, based on the subset of the plurality of temporal parameters, whether the sound data corresponds to the pre-defined sound. 8. The sound identification device of claim 1, wherein to determine whether the sound data corresponds to the pre-defined sound comprises to determine a probability that the sound data corresponds to the pre-defined sound and to compare the probability to a threshold. 9. The sound identification device of claim 1, further comprising a communication module to transmit a notification to a mobile compute device. 10. One or more non-transitory machine-readable media comprising a plurality of instructions stored thereon that, when executed, cause a sound identification device to: acquire sound data;determine a plurality of frames of sound data based on the sound data;determine an acoustic feature matrix having two dimensions and comprising a plurality of first-dimension vectors and a plurality of second-dimension vectors, wherein each second-dimension vector of the plurality of second-dimension vectors corresponds to a corresponding frame of the plurality of frames and each first-dimension vector of the plurality of first-dimension vectors comprises an acoustic feature vector associated with the corresponding frame, andwherein each first-dimension vector of the plurality of first-dimension vectors is associated with a different acoustic feature;determine a plurality of temporal parameters for each first-dimension vector of the plurality of first-dimension vectors;determine based on the pluralities of temporal parameters, whether the sound data corresponds to a pre-defined sound. 11. The one or more non-transitory computer-readable media of claim 10, wherein each acoustic feature vector of the acoustic feature matrix comprises mel-frequency cepstrum coefficients. 12. The one or more non-transitory computer-readable media of claim 10, wherein the pre-defined sound is a cry of an infant. 13. The one or more non-transitory computer-readable media of claim 10, wherein the pre-defined sound is a cough. 14. The one or more non-transitory computer-readable media of claim 10, wherein to determine the plurality of temporal parameters for each first-dimension vector of the plurality of first-dimension vectors comprises to determine the plurality of temporal parameters for each first-dimension vector of the plurality of first-dimension vectors by performing a Fourier-related transform on the corresponding first-dimension vector. 15. The one or more non-transitory computer-readable media of claim 14, wherein the plurality of instructions further cause the sound identification device to select a subset of the plurality of temporal parameters, wherein to determine, based on the plurality of temporal parameters, whether the sound data corresponds to the pre-defined sound comprises to determine, based on the subset of the plurality of temporal parameters, whether the sound data corresponds to the pre-defined sound. 16. The one or more non-transitory computer-readable media of claim 10, wherein to determine, based on the plurality of temporal parameters, whether the sound data corresponds to the pre-defined sound comprises to determine a probability that the sound data corresponds to the pre-defined sound and to compare the probability to a threshold. 17. The one or more non-transitory computer-readable media of claim 10, wherein the plurality of instructions further cause the sound identification device to transmit a notification to a mobile compute device. 18. A method for sound identification by a sound identification device, the method comprising: acquiring, by the sound identification device, sound data;determining, by the sound identification device, a plurality of frames of sound data based on the sound data;determining, by the sound identification device, an acoustic feature matrix having two dimensions and comprising a plurality of first-dimension vectors and a plurality of second-dimension vectors, wherein each second-dimension vector of the plurality of second-dimension vectors corresponds to a corresponding frame of the plurality of frames and each first-dimension vector of the plurality of first-dimension vectors comprises an acoustic feature vector associated with the corresponding frame, andwherein each first-dimension vector of the plurality of first-dimension vectors is associated with a different acoustic feature;determining, by the sound identification device, a plurality of temporal parameters for each first-dimension vector of the plurality of first-dimension vectors;determining, by the sound identification device, based on the pluralities of temporal parameters, whether the sound data corresponds to a pre-defined sound. 19. The method of claim 18, wherein each acoustic feature vector of the acoustic feature matrix comprises mel-frequency cepstrum coefficients. 20. The method of claim 18, wherein the pre-defined sound is a cry of an infant. 21. The method of claim 18, wherein the pre-defined sound is a cough. 22. The method of claim 18, wherein determining the plurality of temporal parameters for each first-dimension vector of the plurality of first-dimension vectors comprises determining the plurality of temporal parameters for each first-dimension vector of the plurality of first-dimension vectors by performing a Fourier-related transform on the corresponding first-dimension vector. 23. The method of claim 22, further comprising selecting, by the sound identification device, a subset of the plurality of temporal parameters, wherein determining, based on the plurality of temporal parameters, whether the sound data corresponds to the pre-defined sound comprises determining, based on the subset of the plurality of temporal parameters, whether the sound data corresponds to the pre-defined sound. 24. The method of claim 18, wherein determining, based on the plurality of temporal parameters, whether the sound data corresponds to the pre-defined sound comprises determining a probability that the sound data corresponds to the pre-defined sound and comparing the probability to a threshold. 25. The method of claim 18, further comprising transmitting a notification to a mobile compute device.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (1)
Hsieh Chau-Kai (Chiung Lin TWX), Baby cry recognizer.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.