Multisensory Speech Detection Using Device Pose and Orientation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mobile computing devices face challenges in accurately detecting and recording speech due to small input devices, difficulty in distinguishing speech from background noise, and issues with initiating recording at the right time, leading to frustrating user experiences.
Innovation Solution
A multisensory speech detection system that uses sensors like accelerometers and proximity sensors to determine the device's position relative to the user, initiating or stopping voice recording based on predefined poses, thereby ensuring accurate capture of speech input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If speech detection is used to enable hands-free interaction, then ease of operation is improved, but reliability deteriorates due to difficulty in distinguishing speech from background noise
Solution Approach 1:
The patent segments the speech detection function into multiple independent components: pose detection (via accelerometers and proximity sensors), speech activation detection, and speech deactivation detection. Each component operates independently to detect specific conditions, and their combined output determines when recording should start and stop. This segmentation allows the system to reliably distinguish user speech from background noise by requiring multiple conditions to be met simultaneously.
2Ease of operation
If automatic speech recording is initiated based on pose detection, then ease of operation is improved, but loss of time worsens due to delays in capturing the beginning of speech
Solution Approach 1:
The patent implements preliminary action by continuously monitoring pose conditions and pre-positioning the recording system to activate immediately when speech conditions are detected. The pose detection system continuously tracks device orientation and proximity, so when the device is held in the correct position near the user's mouth, the system is already prepared to capture speech instantly upon detection, eliminating startup delays and ensuring no portion of the speech is missed.
3Measurement precision
If multiple sensors are used for pose detection, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent merges multiple sensor functions into a unified pose detection system. The accelerometer and proximity sensor data are combined and processed together to determine device pose, rather than operating as separate independent systems. This merging allows the system to achieve high measurement precision for device position and orientation while managing complexity through integrated processing logic that evaluates multiple sensor inputs as a cohesive unit for speech activation decisions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method of multisensory speech detection is disclosed. The method comprises determining an orientation of a mobile device and determining an operating mode of the mobile device based on the orientation of the mobile device. The method further includes identifying speech detection parameters that specify when speech detection begins or ends based on the determined operating mode and detecting speech from a user of the mobile device based on the speech detection parameters.