Microsoft patents method to hide voice identity in speech data
Microsoft describes a system that takes a user's voice recording and removes specific details that reveal their identity. It creates new, synthetic voice data from the text of the recording to help train speech recognition software without using the original, identifiable audio.
In the filing’s own words
A method, computer program product, and computing system for receiving a speech signal from a single microphone. A sensitive speech component is identified from the speech signal. In response to identifying the sensitive speech component, a filtered speech signal is generated by removing the sensitive speech component from the speech signal. A voice style transfer of the speech signal is generated. Speech processing is performed on the filtered speech signal and the voice style transfer of the speech signal.
What exactly they claim (claim 1)
1. A method comprising: receiving, by a speech processing system, a speech signal containing voice characteristics of a user; generating, from the speech signal, training data for an automatic speech recognition (ASR) model by filtering the speech signal to obtain a filtered speech signal and generating a synthetic speech signal from a transcription of the speech signal, the training data including both the filtered speech signal and the synthetic speech signal, wherein the voice characteristics of the user are anonymized in both the filtered speech signal and the synthetic speech signal; and outputting the training data to a device, the training data being used to train the ASR model to perform speech recognition.
Facts
- Company
- Microsoft (Microsoft Technology Licensing, LLC)
- Official title
- SYSTEM AND METHOD FOR MULTI-CHANNEL SPEECH PRIVACY PROCESSING
- Publication number
- US20260311677A1
- Application number
- 19697519
- Filed
- Jun 3, 2026
- Published
- Oct 8, 2026
- Inventors
- 1
- Claims
- 20
Classified under
G10L 13/02 Physics & computingG10L 15/14 Physics & computingMore from Microsoft in Security & privacy
- Microsoft patents system to call 911 for others during video calls Sep 22, 2026
- Microsoft patents system to block fake code suggested by AI Sep 15, 2026
- Microsoft patents one-way data flow to block network attacks Sep 10, 2026
- Microsoft patents voice coding that hides speaker identity for secure calls Sep 17, 2026
- Microsoft patents security checks for AI chats using user and model profiles 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.