Microsoft technology licensing, llc (20240135949). Joint Acoustic Echo Cancellation (AEC) and Personalized Noise Suppression (PNS) simplified abstract

Joint Acoustic Echo Cancellation (AEC) and Personalized Noise Suppression (PNS)

Organization Name

microsoft technology licensing, llc

Inventor(s)

Sefik Emre Eskimez of Bellevue WA (US)

Takuya Yoshioka of Bellevue WA (US)

Huaming Wang of Clyde Hill WA (US)

Alex Chenzhi Ju of Seattle WA (US)

Min Tang of Redmond WA (US)

[[:Category:Tanel P�rnamaa of Tallinn (EE)|Tanel P�rnamaa of Tallinn (EE)]][[Category:Tanel P�rnamaa of Tallinn (EE)]]

Joint Acoustic Echo Cancellation (AEC) and Personalized Noise Suppression (PNS) - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240135949 titled 'Joint Acoustic Echo Cancellation (AEC) and Personalized Noise Suppression (PNS)

Simplified Explanation

The patent application describes a data processing system that utilizes machine learning to perform personalized noise suppression and acoustic echo cancellation in online communication sessions.

• The system receives signals from two computing devices participating in an online communication session: a far-end signal from the first device and a near-end signal from the second device.
• The near-end signal includes speech from a target speaker, an interfering speaker, and an echo signal.
• The system provides the far-end signal, near-end signal, and an indication of the target speaker as input to a machine learning model.
• The machine learning model is trained to analyze the signals and remove speech from interfering speakers and echoes, outputting an audio signal comprising the speech of the target speaker.

---

1. 1. 1. Potential Applications

This technology can be applied in video conferencing platforms, telecommunication systems, and virtual meetings to enhance audio quality and improve communication experiences.

1. 1. 1. Problems Solved

This technology addresses the issue of background noise and echoes during online communication sessions, ensuring clear and focused audio for participants.

1. 1. 1. Benefits

The system improves the overall audio quality of online communication sessions by suppressing noise from interfering speakers and canceling echoes, leading to better understanding and engagement among participants.

1. 1. 1. Potential Commercial Applications

• Virtual meeting platforms
• Telecommunication companies
• Audio conferencing systems

---

1. 1. 1. Possible Prior Art

One potential prior art could be traditional noise suppression and echo cancellation techniques used in audio processing systems before the integration of machine learning algorithms for personalized noise suppression.

1. 1. 1. Unanswered Questions

1. 1. 1. 1. How does the system handle multiple interfering speakers in the near-end signal?

The system is trained to identify and suppress speech from one or more interfering speakers based on the input signals, but the specific mechanisms for handling multiple interfering speakers simultaneously are not detailed in the abstract.

1. 1. 1. 1. What is the computational complexity of the machine learning model for real-time applications?

While the abstract mentions the use of a machine learning model for noise suppression and echo cancellation, it does not provide information on the computational requirements or efficiency of the model for real-time processing in online communication sessions.

Original Abstract Submitted

a data processing system implements receiving a far-end signal associated with a first computing device participating in an online communication session and receiving a near-end signal associated with a second computing device participating in the online communication session. the near-end signal includes speech of a target speaker, a first interfering speaker, and an echo signal. the system further implements providing the far-end signal, the near-end signal, and an indication of the target speaker as an input to a machine learning model. the machine learning model trained to analyze the far-end signal and the near-end signal to perform personalized noise suppression (pns) to remove speech from one or more interfering speakers and acoustic echo cancellation (aec) to remove echoes. the model is trained to output an audio signal comprising speech of the target speaker. the system obtains the audio signal comprising the speech of the target speaker from the model.