17849187. DYNAMIC SPEECH ENHANCEMENT COMPONENT OPTIMIZATION simplified abstract (Microsoft Technology Licensing, LLC)
DYNAMIC SPEECH ENHANCEMENT COMPONENT OPTIMIZATION
Organization Name
Microsoft Technology Licensing, LLC
Inventor(s)
Ross G. Cutler of Clyde Hill WA (US)
William D. Fallas Cordero of Belen (CR)
DYNAMIC SPEECH ENHANCEMENT COMPONENT OPTIMIZATION - A simplified explanation of the abstract
This abstract first appeared for US patent application 17849187 titled 'DYNAMIC SPEECH ENHANCEMENT COMPONENT OPTIMIZATION
Simplified Explanation
The patent application describes a method for optimizing speech enhancement components in speech communication systems using non-intrusive speech quality assessment.
- The method involves receiving audio data that includes speech and has been processed by speech enhancement components.
- A trained non-intrusive speech quality assessment (NISQA) model is used to detect the quality of the speech in the audio data automatically.
- Based on the detected quality of the speech, one or more of the speech enhancement components are changed.
Potential Applications
- Speech communication systems
- Telecommunication systems
- Voice assistants
- Audio recording and playback devices
Problems Solved
- Inefficient speech enhancement components in speech communication systems
- Difficulty in determining the quality of speech automatically
- Lack of optimization in speech enhancement based on speech quality
Benefits
- Improved speech quality in communication systems
- Automatic detection and optimization of speech enhancement components
- Enhanced user experience in speech communication
- Efficient utilization of speech enhancement technology
Original Abstract Submitted
Systems, methods, and computer-readable storage devices are disclosed for optimizing speech enhancement components to use in speech communication systems using non-intrusive speech quality assessment. One method including: receiving audio data, the audio data including speech; and the audio data having been processed by at least one speech enhancement component; detecting a first quality of the speech of the audio data using a trained non-intrusive speech quality assessment (NISQA) model, the trained NISQA model trained to detect quality of speech automatically; and changing one or more of the at least one speech enhancement component based on the detected first quality of the speech.