Apple's cutting-edge technology! AirPods receive a new patent: it can monitor the user's breathing rate.

Respiration rate (RR) is an important physiological indicator for measuring human health, usually expressed as breaths per minute (BPM). Accurate measurement of respiratory rate is crucial in fields such as medical monitoring, sports health, and sleep analysis. However, traditional respiratory rate detection methods (such as chest strap sensors and nasal airflow monitoring) often rely on specialized hardware, which suffers from inconvenience and low comfort.

On May 1, 2025, Apple published a patent titled "Determining Breathing Rate Based on Audio Stream and User Status" (US20250134408A1), which aims to achieve respiratory monitoring using the existing hardware architecture of head-mounted devices such as AirPods.

Technical Overview

This patent describes a system and method for determining respiratory rate using an audio stream and user status. The system receives input indicating the user's status, as well as audio streams from a built-in microphone (in-ear microphone) and an external microphone of a head-mounted device. Depending on the user's status, the system selectively utilizes the internal and/or external audio streams to determine the user's respiratory rate. In some cases, the system invokes a machine learning model to extract breathing signals from the audio streams based on the user's status. This method is applicable to various scenarios (such as meditation or exercise) and improves the accuracy of respiratory rate detection by adjusting the weights of the internal and external audio streams.

Core content

1. Dual-microphone signal fusion: Breathing signals are collected simultaneously using an in-ear microphone (internal audio stream) and an external microphone (external audio stream) to improve the signal-to-noise ratio.

2. User Status Awareness: Dynamically adjust signal processing strategies based on the user's current activity status (such as exercise, meditation, rest) to optimize respiratory rate calculation.

3. Machine learning-assisted enhancement: Blind source separation and deep learning models are used to extract respiratory signals, reducing environmental noise interference.

System Architecture

1. Hardware Components

(1) Head-mounted devices

• In- ear microphone : Collects breathing sounds (such as exhalation/inhalation airflow) inside the ear canal.
• External microphone : Collects ambient sound to assist in noise reduction or breathing detection during exercise.
• Speaker : Used to play audio (such as music, calls), but may introduce interference and requires echo cancellation.

(2) Supporting equipment

Smartphones, smartwatches, and other devices are used to receive user input (such as sports modes) or provide additional sensor data (such as heart rate and motion acceleration).

2. Signal Processing Flow

(1) Signal acquisition

In-ear microphone → Internal audio stream

External microphone → External audio stream

(2) Preprocessing (filtering & enhancement)

Low-pass filter (LPF) : preserves respiratory-related frequencies (typically 0.1–2 kHz).

Blind source separation (BSS) : Separating respiratory signals from noise (such as music and ambient sound).

Echo cancellation (AEC) : If the device is playing audio (such as music), this interference needs to be removed.

(3) Feature extraction

Time-frequency analysis (Spectrogram/MFCC): Detects the respiratory cycle (inspiration/expiration)

Machine learning model: A model consisting of a two-layer GRU recurrent network and a fully connected layer is used to extract respiratory cycle features through time series analysis.

• Input: Audio features + user status (e.g., exercise, meditation).
• Output: Estimated respiratory rate.

(4) Dynamic weight adjustment

Adjust the contribution weights of internal/external audio streams based on user status.

• Meditation Mode: Primarily relies on internal audio streams (clearer signals inside the ear).
• Motion mode: Primarily relies on external audio streams (motion noise affects the signal inside the ear).

(5) Post-processing & verification

Trust score: assesses the reliability of respiratory rate calculation.

Multi-sensor fusion: Combining heart rate and motion data (such as smartwatches) for cross-validation.

Technological Innovation Points

1. Dual-microphone collaboration and blind source separation

AirPods use an internal microphone to collect breathing sounds from the ear canal and an external microphone to capture ambient noise. Combined with blind source separation technology, they filter out interference from music, human voices, and other sources to directly extract the breathing signal. This design maintains data accuracy even in complex environments such as subways and offices.

2. Multimodal data fusion

The system dynamically adjusts its algorithm based on exercise type and heart rate data, for example, by increasing the respiratory rate threshold during exercise to avoid misjudgments. The machine learning model can adapt to the breathing characteristics of different users, reducing the impact of individual differences.

3. Expanding Scenario-Based Applications

Breathing rate is closely related to cardiopulmonary function and metabolic status; for example, rapid breathing may indicate an asthma attack or altitude sickness. The portability of AirPods makes them suitable for daily health management, such as guiding breathing rhythm during exercise and providing early warnings of sleep apnea. Furthermore, combining them with blood oxygen and heart rate data from Apple Watch allows for the creation of a more comprehensive health profile.

Summarize

Currently, no product in the consumer market adopts the exact same technological approach as this patent. Manufacturers like Huawei and Samsung primarily focus on HRV and motion monitoring, while there is a significant gap between medical devices and consumer products. Technical challenges include signal separation in complex noise environments, power consumption control during long-term monitoring, and the algorithm's adaptability to different breathing patterns. If Apple can translate this patent into practical functionality, it could redefine the health monitoring capabilities of headphones.

This patent reflects the trend of wearable devices moving towards "non-invasive monitoring"-achieving non-invasive collection of physiological indicators through everyday devices (such as headphones and glasses). In the future, audio analysis technology may further expand into areas such as voice emotion recognition and stress monitoring, promoting the deep integration of health technology and consumer electronics.

Note: This article is reprinted from 21dB Acoustics.