Dynamic VoiceXML Grammar Baseform Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
VoiceXML currently lacks the ability to support applications where grammars can be built with dynamically generated acoustic baseforms from user utterances, limiting the functionality of voice-enabled applications such as phone dialers.
Innovation Solution
Introducing a new element to the VoiceXML language with specific attributes that enable the generation of acoustic baseforms from user recordings, allowing for dynamic grammar rule creation and binding within the VoiceXML platform.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If VoiceXML uses static grammars generated from text data, then the grammar structure is simple and maintainable, but the system cannot dynamically adapt to user-specific utterances and acoustic variations
Solution Approach 1:
The patent introduces dynamic grammar generation where the grammar rules are not fixed but are generated in real-time based on recorded user utterances. The system dynamically creates acoustic baseforms and grammar rules from actual user speech patterns, allowing the grammar to adapt to individual users' pronunciation and speech characteristics.
Solution Approach 2:
The system enables self-service grammar generation where the grammar is automatically created from user utterances without requiring manual programming of grammar rules. The ASR system automatically extracts acoustic baseforms and generates appropriate grammar rules from the recorded speech, allowing the system to serve itself in creating customized grammars.
2Measurement precision
If the system records and processes user utterances to generate dynamic baseforms, then the grammar becomes more accurate for user-specific commands, but the processing time and computational resources increase
Solution Approach 1:
The system performs preliminary recording of user utterances during initial interaction, then generates the grammar rules in advance before actual command processing. This preliminary grammar generation allows for accurate recognition while reducing the processing time during subsequent command execution, as the grammar structure is already prepared.
Solution Approach 2:
The system continuously processes user utterances and updates the grammar rules in real-time as users interact with the system. This continuous generation and updating of grammar rules ensures that the system always has the most accurate and up-to-date understanding of user speech patterns, improving recognition accuracy without significant time penalties.
3Adaptability or versatility
If VoiceXML adds new elements for dynamic baseform generation, then the system gains capability for adaptive grammar creation, but the language complexity and parsing difficulty increase
Solution Approach 1:
The patent introduces nested structural elements where the new dynamic baseform generation capabilities are embedded within the existing VoiceXML framework. The new elements are nested within standard VoiceXML constructs, allowing the dynamic grammar creation functionality to be integrated without requiring complete redesign of the language structure.
Solution Approach 2:
The new elements introduced for dynamic baseform generation are designed to serve multiple functions within the VoiceXML system. These elements can be used for both grammar generation and for defining acoustic characteristics, reducing the need for separate dedicated elements and thereby limiting the increase in language complexity.
Data Source
AI summary
A method (200) of building a grammar with baseforms generated dynamically from user utterances can include the steps of recording (205) a user utterance, generating (210) a baseform using the user utterance, creating or adding to (215) a grammar rule using the baseform, and binding (230) the grammar rule in a grammar document of a voice extensible markup language program. Generating a baseform can optionally include introducing a new element to VoiceXML with attributes that enable generating the baseform from a referenced recording such as the user utterance. In one embodiment, the method can be used to create (235) a phonebook and a grammar to access the phonebook by repeatedly visiting a form containing the grammar rule with attributes that enable generating the baseform from the referenced recording.