20230162726. OBFUSCATING AUDIO SAMPLES FOR HEALTH PRIVACY CONTEXTS simplified abstract (International Business Machines Corporation)
OBFUSCATING AUDIO SAMPLES FOR HEALTH PRIVACY CONTEXTS
Organization Name
International Business Machines Corporation
Inventor(s)
Victor Abayomi Akinwande of Karen (KE)
Komminist Weldemariam of Ottawa (CA)
OBFUSCATING AUDIO SAMPLES FOR HEALTH PRIVACY CONTEXTS - A simplified explanation of the abstract
This abstract first appeared for US patent application 20230162726 titled 'OBFUSCATING AUDIO SAMPLES FOR HEALTH PRIVACY CONTEXTS
Simplified Explanation
The patent application describes a system for detecting and removing bio-markers from audio samples using machine learning techniques. The system consists of a supervised discriminator, a denoising autoencoder, and a conditional auxiliary generative adversarial network (GAN).
- The supervised discriminator is trained to identify bio-markers in audio samples.
- The denoising autoencoder learns a latent space that can reconstruct an output audio sample with the same quality as the input sample.
- The conditional auxiliary GAN is trained to generate an output audio sample without the bio-markers, while maintaining the fidelity of the input sample.
- The system can be deployed in an audio processing system to detect and remove bio-markers from audio samples.
Potential Applications
- Audio forensics: The system can be used to analyze audio recordings for the presence of bio-markers, such as breathing sounds or heartbeats, which can be useful in forensic investigations.
- Healthcare: The system can be applied in healthcare settings to analyze audio samples for the presence of specific bio-markers, aiding in the diagnosis of medical conditions.
- Noise reduction: The denoising autoencoder can be used to remove unwanted background noise from audio recordings, improving the overall audio quality.
Problems Solved
- Detection of bio-markers: The system provides a method for accurately detecting bio-markers in audio samples, which can be challenging due to variations in recording conditions and background noise.
- Bio-marker removal: The system offers a solution for removing bio-markers from audio samples, allowing for clearer analysis and processing of the audio data.
Benefits
- Accuracy: The supervised discriminator and conditional auxiliary GAN improve the accuracy of bio-marker detection and removal, ensuring reliable results.
- Fidelity preservation: The denoising autoencoder reconstructs the output audio sample with the same fidelity as the input sample, maintaining the quality of the audio data.
- Versatility: The system can be applied to various audio processing tasks, including forensics, healthcare, and noise reduction, making it a versatile tool for different industries.
Original Abstract Submitted
a supervised discriminator for detecting bio-markers in an audio sample dataset is trained and a denoising autoencoder is trained to learn a latent space that is used to reconstruct an output audio sample with a same fidelity as an input audio sample of the audio sample dataset. a conditional auxiliary generative adversarial network (gan) trained to generate the output audio sample with the same fidelity as the input audio sample, wherein the output audio sample is void of the bio-markers. the conditional auxiliary generative adversarial network (gan), the corresponding supervised discriminator, and the corresponding denoising autoencoder are deployed in an audio processing system
