Verbal Record Analysis System Using Phonetic Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automatic speech recognition (ASR) systems fail to accurately convert verbal dictations into concise written formats due to issues like disfluencies, omission of function words, and specific formatting conventions essential in medical transcription, leading to significant differences between initial ASR output and final documents.
Innovation Solution
A system comprising a database and processor that analyzes verbal records, extracts relevant features such as acoustic and phonetic features, and applies reasoning approaches like heuristic algorithms to generate processed records, including punctuation, headings, and capitalization, converting them into formats like XML.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional ASR is used to convert verbal dictation to text, then the basic transcription function is provided, but the output lacks accurate formatting, punctuation, and structural conventions required for medical documentation
Solution Approach 1:
The system segments the ASR processing into distinct modules: acoustic feature extraction, phonetic feature extraction, reasoning approach application, and formatting generation. Each module handles specific aspects of transcription accuracy, allowing the complex overall task to be managed through coordinated simpler components.
Solution Approach 2:
The system performs preliminary extraction of acoustic and phonetic features from the verbal record before applying reasoning approaches to generate formatted output. This preliminary processing prepares the data in advance, enabling more accurate and convention-compliant transcription without requiring complex real-time processing.
2Productivity
If ASR systems process verbal records quickly, then productivity is improved, but the quality of transcription and formatting deteriorates
Solution Approach 1:
By pre-extracting acoustic and phonetic features and pre-processing the verbal records, the system enables faster subsequent processing. The reasoning approaches can operate on prepared feature data rather than raw audio, maintaining high transcription quality while improving overall processing speed.
Solution Approach 2:
The system maintains continuous processing through automated feature extraction and reasoning application, eliminating manual intervention steps. This continuous automated action ensures consistent high-quality transcription while maintaining high productivity through uninterrupted processing flow.
3Measurement precision
If the system extracts and analyzes multiple verbal features, then the accuracy of processed records improves, but the computational complexity and processing time increase
Solution Approach 1:
The computational complexity is segmented into distinct feature extraction stages: acoustic feature extraction, phonetic feature extraction, and reasoning approach application. Each stage processes specific types of features independently, making the overall complex computation manageable through modular processing.
Solution Approach 2:
The system transforms the verbal record into different parameter representations (acoustic features, phonetic features) through systematic parameter changes. This transformation enables precise feature analysis while managing computational complexity through standardized parameter extraction processes.
Data Source
AI summary
Language dictation recognition systems and methods for using the same. In at least one exemplary system for analyzing verbal records, the system comprises a database capable of receiving a plurality of verbal records, the verbal record comprising at least one identifier and at least one verbal feature and a processor operably coupled to the database, where the processor has and executes a software program. The processor being operational to identify a subset of the plurality of verbal records from the database, extract at least one verbal feature from the identified records, analyze the at least one verbal feature of the subset of the plurality of verbal records, process the subset of the plurality of records using the analyzed feature according to at least one reasoning approach, generate a processed verbal record using the processed subset of the plurality of records, and deliver the processed verbal record to a recipient.