Microsoft patents audio encoding method using modulation domains and ML
Microsoft describes a system that converts voice audio signals into a 'modulation domain' to encode them using specific carrier and modulator signals. The encoded data is then converted back to the time domain and processed by a machine learning model to reconstruct the original voice signal.
In the filing’s own words
A method, computer program product, and computing system for processing an audio signal by converting the audio signal to the modulation domain. The modulation domain audio signal is encoded with a plurality of carrier signals and a plurality of modulator signals derived from the modulation domain audio signal. The encoded modulation domain audio signal is converted to the time domain.
What exactly they claim (claim 1)
1. A method comprising: receiving an encoded voice audio signal from a transmitter via a telecommunications network; and decoding the encoded voice audio signal using a machine learning (ML) model to obtain a decoded voice audio signal, wherein decoding the encoded voice audio signal includes: converting the encoded voice audio signal into the modulation domain to obtain a modulation-domain representation of the encoded voice audio signal; decoding the modulation-domain representation of the encoded voice audio signal; converting the decoded modulation-domain representation of the encoded voice audio signal to the time domain to obtain a time-domain representation of the encoded voice audio signal; and processing the time-domain representation of the encoded voice audio signal to obtain the decoded voice audio signal; and outputting the decoded voice audio signal to an automated speech recognition (ASR) system.
Facts
- Company
- Microsoft (Microsoft Technology Licensing, LLC)
- Official title
- SYSTEM AND METHOD FOR MODULATION DOMAIN-BASED AUDIO SIGNAL ENCODING
- Publication number
- US20260279380A1
- Application number
- 19671217
- Filed
- May 7, 2026
- Published
- Sep 17, 2026
- Inventors
- 3
- Claims
- 20
Classified under
G10L 21/04 Physics & computingMore from Microsoft in Audio & voice
- Microsoft patents AI model that cleans up video calls by removing echoes and background voices Oct 1, 2026
- Microsoft patents system to grade how well audio fills in missing data Sep 29, 2026
- Microsoft patents audio effect that turns group speech into background chatter Sep 29, 2026
- Microsoft patents audio tool that adjusts sound based on hearing tests Sep 29, 2026
- Microsoft patents audio system that removes private details from recordings Oct 1, 2026
Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.