Sleep inducing device
A sleep inducing device is disclosed. The sleep inducing device according to an embodiment of the present invention comprises: a sound output part for outputting a sound; a sensing part for acquiring a respiration signal of a user; and a processor which acquires a respiration frequency of the user on the basis of the respiration signal, determines a characteristic frequency of a respiration conforming sound on the basis of the respiration frequency, and outputs the respiration confirming sound according to the characteristic frequency.
1. A sleep induction device comprising:
a speaker configured to output sound;
a sensor configured to obtain a breathing signal of a user; and
a processor configured to:
obtain a breathing frequency of the user per a first time period based on the obtained breathing signal;
determine and generate a breathing-tuned sound according to a characteristic frequency by combining a breathing induction signal with an ambient noise masking sound, wherein the characteristic frequency of the breathing-tuned sound including the ambient noise masking sound is determined according to a characteristic of the breathing induction signal; and
output the breathing-tuned sound for inducing sleep in the user by inducing the user's breathing while simultaneously reducing the ambient noise,
wherein the characteristic of the breathing induction signal corresponds to a magnitude of the obtained breathing signal or a frequency of the obtained breathing signal for a second time period.
2. The sleep induction device of claim 1 , wherein the characteristic frequency of the breathing tuned sound is less than the obtained breathing frequency.
3. The sleep induction device of claim 2 , wherein a difference between the characteristic frequency and the obtained breathing frequency is maintained at a preset value.
4. The sleep induction device of claim 1 , wherein the processor is further configured to:
output the breathing-tuned sound according to the characteristic frequency being a first frequency,
reduce the frequency of the breathing-tuned sound from the first frequency to a second frequency based on the obtained breathing frequency, wherein the second frequency is less than the first frequency, and
output the breathing-tuned sound according to the second frequency.
5. The sleep induction device of claim 4 , wherein the processor is further configured to determine a point in time at which the characteristic frequency changes from the first frequency to the second frequency based on the obtained breathing frequency.
6. The sleep induction device of claim 5 , wherein
the breathing-tuned sound is output according to the second frequency based on a difference between the obtained breathing frequency and the first frequency becoming smaller than a predetermined magnitude.
7. The sleep induction device of claim 5 , wherein the processor is further configured to
output the breathing-tuned sound according to the second frequency based on the obtained breathing frequency decreasing to the first frequency.
8. The sleep induction device of claim 4 , wherein the processor is further configured to
output the breathing-tuned sound according to the second frequency based on a preset time elapsing since the breathing-tuned sound is output according to the first frequency.
9. The sleep induction device of claim 1 , wherein the processor is further configured to:
provide the obtained breathing frequency as an input to a reinforcement learning model,
output the breathing-tuned sound according to at least one of a point in time at which the characteristic frequency recommended by the reinforcement learning model changes or a delta in the characteristic frequency, and
train the reinforcement learning model using a reaction of the user.
10. The sleep induction device of claim 9 , wherein the reaction of the user comprises at least one of a change in the breathing frequency, a movement of the user, or a time required for the user to enter a sleep state.
11. The sleep induction device of claim 1 , wherein a minimum value of the characteristic frequency corresponds to the breathing frequency obtained during a sleep state of the user.
12. The sleep induction device of claim 1 , wherein the processor is further configured to adjust a maximum magnitude of the breathing-tuned sound based on environment information determined by the processor.
13. The sleep induction device of claim 1 , wherein the processor is further configured to:
provide the breathing frequency and environment information as an input to a reinforcement learning model,
output the breathing-tuned sound according to at least one of a point in time at which the characteristic frequency recommended by the reinforcement learning model changes, a delta in the characteristic frequency, a maximum magnitude of the breathing-tuned sound, or a type of the breathing-tuned sound, and
train the reinforcement learning model using a reaction of the user.
14. A method for operating a sleep induction device, the method comprising:
obtaining a breathing signal of a user by a sensor of the device;
obtaining, by a processor of the device, a breathing frequency of the user per a first time period based on the obtained breathing signal;
determining and generating, by the processor, a breathing-tuned sound according to a characteristic frequency by combining a breathing induction signal with an ambient noise masking sound, wherein the characteristic frequency of the breathing-tuned sound including the ambient noise masking sound corresponds to a characteristic of the breathing induction signal; and
outputting, by a speaker of the device, the breathing-tuned sound for inducing sleep in the user by inducing the user's breathing while simultaneously reducing the ambient noise,
wherein the characteristic of the breathing induction signal corresponds to a magnitude of the breathing induction signal or a frequency of the breathing induction signal for a second time period.