Real-Time Haptic Feedback via Audio Event Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing haptic feedback technologies in mobile electronic products lack mature applications based on event detection and often require high audio quality, are limited to single use scenarios, and provide poor user experience due to inadequate vibration matching for audio events.
Innovation Solution
A haptic feedback method that uses an algorithmically trained support vector machine (SVM) model to identify audio event types in real-time, extracting MFCC features from audio clips and matching them with corresponding vibration effects according to a preset rule, enhancing user experience by providing customized haptic feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing haptic feedback schemes are used to match vibrations for audio, then haptic feedback functions can be provided, but high requirements on audio quality are needed and user experience is poor
Solution Approach 1:
The patent introduces audio event detection as an intermediary layer between audio input and haptic feedback output. Instead of directly mapping audio signals to vibrations, the system first detects and classifies audio events (e.g., shooting, explosion, collision) using machine learning models, then triggers appropriate haptic effects based on detected event types. This intermediary processing reduces dependency on high audio quality while improving haptic feedback accuracy and user experience.
2Adaptability or versatility
If existing haptic feedback schemes are used, then vibration matching can be provided, but application scenarios are limited and user experience is poor
Solution Approach 1:
The patent implements a universal audio event detection framework that can identify multiple types of audio events (shooting, explosion, collision, footstep, etc.) across different application scenarios including games, videos, and music. The system uses trained machine learning models to detect various event types and triggers corresponding haptic feedback effects, making the haptic feedback system adaptable and versatile across diverse use cases while maintaining reliable user experience.
3Reliability
If event detection based haptic feedback is implemented, then user experience can be improved, but algorithm processing and model training are required
Solution Approach 1:
The patent applies preliminary action by pre-training machine learning models offline to recognize audio events. The system collects audio data, trains classification models to identify different event types (shooting, explosion, collision, etc.), and stores the trained models for later use. During actual operation, the pre-trained models quickly process audio input to detect events and trigger haptic feedback, reducing real-time processing complexity while maintaining high user experience quality.
Data Source
AI summary
Provided a haptic feedback method, including: step S1 of algorithmically training an audio clip containing a known audio event type to obtain an algorithm model; and step S2 of obtaining an audio, identifying the audio by the algorithm model to obtain different audio event types in this audio, matching, according to a preset rule, the audio event types with different vibration effects as a haptic feedback and outputting the haptic feedback. Compared with the related art, the present haptic feedback method provides users with real-time haptic feedback when applied to a mobile electronic product, thereby achieving excellent use experience of the mobile electronic product.