Multimodal Speech Recognition Grammar for Selection List Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multimodal applications face challenges in determining whether a user wants to select or deselect multiple items using speech input, particularly in scenarios with large selection lists, as existing technologies lack a clear method to anticipate user intentions and control cumulative option selection effectively.
Innovation Solution
A multimodal application with a speech recognition grammar that includes a semantic interpretation script, capable of producing a token indicating whether to select or deselect items, and a configuration parameter to manage user-editable settings for handling item selections, allowing the application to determine the appropriate action based on the token and parameter values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If speaker independent voice recognition is implemented to enhance user experience, then ease of operation is improved, but measurement precision deteriorates due to limited vocabulary constraints
Solution Approach 1:
The system dynamically adapts the grammar structure based on the selection list context. The grammar is not static but is configured to recognize speech in the context of available options, allowing the recognition accuracy to improve while maintaining ease of operation through natural speech input.
Solution Approach 2:
The system changes the parameters of speech recognition by using semantic interpretation tokens that indicate selection or deselection actions. This allows the system to interpret user intent more accurately without constraining the vocabulary, thereby improving measurement precision while maintaining ease of operation.
2Reliability
If semantic interpretation scripts are added to grammar to produce tokens indicating selection actions, then reliability is improved, but device complexity increases
Solution Approach 1:
The grammar is segmented into distinct components: the base grammar for recognizing items and the semantic interpretation script for determining user intent. This segmentation allows each component to be optimized independently, improving reliability while managing complexity through modular design.
Solution Approach 2:
The semantic interpretation token acts as an intermediary between the speech recognition process and the selection action. It mediates the translation from spoken input to intended action (select or deselect), improving reliability by making the interpretation explicit while keeping the grammar structure manageable.
3Adaptability or versatility
If configuration parameters are made user-editable to control cumulative selection behavior, then adaptability is improved, but device complexity increases
Solution Approach 1:
The system performs preliminary configuration by setting default values for the cumulative selection parameter. This allows the system to adapt to user preferences in advance, improving adaptability while reducing the immediate complexity by providing sensible defaults that users can modify if needed.
Data Source
AI summary
Methods, apparatus, and computer program products are described for automatic speech recognition (‘ASR’) that include accepting by the multimodal application speech input and visual input for selecting or deselecting items in a selection list, the speech input enabled by a speech recognition grammar; providing, from the multimodal application to the grammar interpreter, the speech input and the speech recognition grammar; receiving, by the multimodal application from the grammar interpreter, interpretation results including matched words from the grammar that correspond to items in the selection list and a semantic interpretation token that specifies whether to select or deselect items in the selection list; and determining, by the multimodal application in dependence upon the value of the semantic interpretation token, whether to select or deselect items in the selection list that correspond to the matched words.