18207406. Speech Enhancement simplified abstract (Nokia Technologies Oy)
Contents
Speech Enhancement
Organization Name
Inventor(s)
Juha Tapio Vilkamo of Helsinki (FI)
Kai Petteri Havukainen of Lempaala (FI)
Speech Enhancement - A simplified explanation of the abstract
This abstract first appeared for US patent application 18207406 titled 'Speech Enhancement
Simplified Explanation
The patent application is about a speech enhancement technology that can be adjusted for different sound environments. It involves obtaining a control parameter that represents a user's preference for speech enhancement. The audio signals are then processed to determine the type of sound present. The control parameter and sound classification are used to determine a processing parameter. This processing parameter is used to enable speech enhancement on the audio signals, controlling the balance between speech and other sounds in the output signal.
- Obtaining a control parameter representing user preference for speech enhancement
- Processing audio signals to determine sound classification
- Using control parameter and sound classification to determine processing parameter
- Enabling speech enhancement on audio signals using the processing parameter
- Controlling proportions of speech and other sounds in the output signal
Potential Applications
- Telecommunications: Improving speech quality in phone calls or video conferences
- Broadcasting: Enhancing speech clarity in radio or TV broadcasts
- Voice assistants: Improving speech recognition and understanding in smart speakers or virtual assistants
Problems Solved
- Inconsistent speech quality in different sound environments
- Difficulty in distinguishing speech from background noise or other sounds
- Lack of user control over the balance between speech and other sounds
Benefits
- Improved speech clarity and intelligibility
- Customizable speech enhancement based on user preferences
- Enhanced user experience in various audio applications
Original Abstract Submitted
Examples of the disclosure relate to speech enhancement that can be adapted for varying sound scenes. In examples of the disclosure a control parameter for speech enhancement is obtained. The control parameter indicates a user preference for speech enhancement. One or more audio signals are obtained and the one or more audio signals are processed to determine a sound classification based at least on the one or more audio signals. The control parameter and the sound classification are used to determine a processing parameter. Speech enhancement is enabled on the one or more audio signals. The speech enhancement uses the processing parameter such that the processing parameter is configured to control proportions of speech and remainder in an output signal.
