Speech recognition method and system based on large model and speech synthesis engine

By generating high-quality speech training data through large language models and speech synthesis engines, and combining context awareness and semantic disambiguation technology, the problems of low recognition accuracy and high training cost of traditional ASR systems in complex scenarios are solved, achieving efficient and low-cost speech recognition optimization.

CN121306105APending Publication Date: 2026-01-09GUANGZHOU AUSUN INFORMATION TECH CO LTD
0 Cites 0 Cited by

Patent Information

Application Number
CN202511813762.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-12-04
Publication Date
2026-01-09

AI Technical Summary

Technical Problem

Traditional ASR systems have low recognition accuracy in complex scenarios such as accents, dialects, technical terms, background noise, and multi-person conversations. They also have difficulty distinguishing homophones and near-homophones and lack high-quality speech training data, resulting in long iteration cycles and high costs.

Method used

By employing a large language model and a speech synthesis engine, high-quality speech training data is generated. Combined with context-aware and semantic disambiguation techniques, automatic speech recognition training and verification are performed to form a training closed loop and continuously improve the model.

Benefits of technology

It achieves high-accuracy speech recognition in complex scenarios, reduces training costs, and improves the model's self-optimization ability and recognition accuracy.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of speech recognition, provides a speech recognition method and system based on a large model and a speech synthesis engine, and greatly improves the recognition accuracy of ASR recognition by combining a large language model language with a new generation of speech synthesis engine. A large language model is utilized to generate a large batch of high-quality corpora, and a speech synthesis engine is utilized to obtain a large amount of high-quality speech training data for training an automatic speech recognition model and checking and correcting a recognition result at the same time. A training closed loop of generating high-quality text training data, synthesizing natural speech, recognizing natural speech, finding errors, correcting the errors and generating training data is realized, so that an automatic speech recognition model can continuously, professionally and automatically reinforce learning at low cost; and furthermore, a high-accuracy and high-quality speech recognition result is output under semantic errors and complex scenes.
Need to check novelty before this filing date? Find Prior Art