Personal Audio Singing Detection Using Bone-Conducted Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing personal audio devices struggle to differentiate between a user speaking and singing, leading to suboptimal operation such as volume adjustment and noise cancellation, which can detract from the user experience.
Innovation Solution
The method involves using bone-conducted speech signals from a transducer and output audio signals to determine if a user is singing, adapting operations like ANC, passthrough, and voice assistance based on these signals, utilizing metrics such as speech articulation rate, peak modulation frequency, and correlation between audio signals to make accurate determinations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If active processing is adapted when the user is speaking, then conversation facilitation is improved, but user experience deteriorates when the user is singing
Solution Approach 1:
The system dynamically adapts audio processing parameters based on real-time detection of the user's vocal state. By continuously monitoring bone-conducted speech signals and comparing them against playback signals, the system adjusts active processing settings to match the current activity (speaking vs. singing), thereby optimizing performance for each specific use case
Solution Approach 2:
The invention changes processing parameters such as volume level adjustments and noise cancellation settings based on the detected vocal state. When singing is detected, the system modifies these parameters differently than when speaking is detected, ensuring appropriate audio processing for each activity type
2Ease of operation
If voice assistant is enabled for normal use, then user interaction is improved, but false triggers occur during singing
Solution Approach 1:
The system replaces traditional voice-activated triggers with a more sophisticated detection mechanism that analyzes bone-conducted speech signals. By examining characteristics such as articulation rate and correlation with playback signals, the system can distinguish between intentional voice assistant activation and singing, preventing false triggers while maintaining normal voice interaction capabilities
3Object-affected harmful factors
If noise cancellation is applied during speech, then background noise reduction is improved, but singing quality deteriorates
Solution Approach 1:
The system dynamically adjusts noise cancellation settings based on the detected vocal state. When singing is detected through analysis of bone-conducted speech characteristics, the system modifies or disables noise cancellation processing to preserve singing quality, while maintaining effective noise reduction during normal speech activities
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances user experience by optimizing audio device operations during singing, reducing noise interference and preventing false triggers of voice assistants, thus improving sound quality and user interaction.
Implementation Method 1
receiving a first audio signal comprising bone-conducted speech of the user from a first transducer
Data Source
AI summary
A method of detecting singing of a user of a personal audio device, the method comprising: receiving a first audio signal comprising bone-conducted speech of the user from a first transducer of the personal audio device; monitoring a second audio signal output to a speaker of the personal audio device; and determining whether the user is singing based on the first audio signal and the second audio signal.