Real-Time Large Language Speech-to-Text Segmentation and Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speech to text systems are computationally expensive and introduce latency, and splitting communications into segments leads to a loss of accuracy due to their design limitations.
Innovation Solution
The system splits conversations into smaller chunks for parallel processing using transformer neural networks and employs a domain-specific correction module to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the communication is processed as a whole by legacy speech to text systems, then transcription accuracy is maintained, but computational cost increases and latency is introduced
Solution Approach 1:
The patent divides the communication into multiple overlapping segments or chunks, allowing parallel processing of each segment while maintaining the ability to reconstruct the complete transcription. This segmentation enables faster processing through parallel computation while managing accuracy through combination strategies.
2Productivity
If the communication is split into multiple segments for parallel processing, then processing speed increases and resource consumption decreases, but transcription accuracy deteriorates
Solution Approach 1:
The patent combines multiple transcribed segments into a complete transcription by leveraging overlapping regions and contextual information. The combination process reconciles discrepancies between segments to produce an accurate final transcription, effectively merging the benefits of parallel processing with the accuracy of whole-communication processing.
Solution Approach 2:
The system performs preliminary transcription on individual segments in parallel, then applies correction models to fix errors before final combination. This preliminary processing allows errors to be identified and corrected systematically, maintaining accuracy while benefiting from parallel computation.
3Measurement precision
If legacy speech to text systems process entire conversations, then phrase-level accuracy is achieved, but computational resources are excessively consumed
Solution Approach 1:
The communication is divided into smaller segments that can be processed independently and in parallel, reducing the computational burden on each processing unit. This segmentation allows modern systems to handle large communications efficiently while maintaining accuracy through combination of segment results.
Data Source
AI summary
Methods and systems for transcribing communications are provided. Methods may include receiving a communication. Methods may include splitting the communication into a plurality of communication segments. Each communication segment may include two or more words. Methods may include transcribing each segment included in the plurality of communication segments, in parallel. The transcribing may include using a transformer neural network to transcribe each segment included in the plurality of communication segments. Methods may include generating a transcription from the transcribing. The transcription may be generated by combining the transcription of each of the communication segments into a combined transcription. Methods may include correcting the combined transcription.