Speech recognition error correction method, device and readable medium based on artificial intelligence

Abstract

The invention provides an artificial intelligence-based voice recognition error correction method, device and readable medium. The method includes: determining the target field of speech recognition; when the original text of the speech recognition has an original segment whose fluency is not greater than a preset fluency threshold in the target field, the ED segment is recalled for the original segment that needs error correction, and the original A collection of candidate fragments for fragments; use the pre-trained fragment scoring model to score each candidate fragment; according to the scoring of each candidate fragment, obtain the corresponding target of each original fragment in the target field from the set of candidate fragments by decoding fragments, resulting in corrected text for speech recognition in the target domain. The technical proposal of the present invention makes up for the deficiencies of the prior art, and provides an error correction scheme for speech recognition, which can correct erroneous segments in the original text of speech recognition, and effectively improve the accuracy of speech recognition.

Description

【Technical field】 [0001] The invention relates to the field of computer application technology, in particular to an artificial intelligence-based speech recognition error correction method, device and readable medium. 【Background technique】 [0002] Artificial Intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. Artificial intelligence is a branch of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a manner similar to human intelligence. Research in this field includes robotics, language recognition, image recognition, natural language processing and expert systems, etc. [0003] With the development of smart technology, smart devices can receive the user's voice in various scenarios and recognize the user's voice as text. And obtain the corresponding result ac...

AI Technical Summary

Problems solved by technology

[0004] However, in the existing technical solutions above, the smart device does not correct the recognized speech recognition text, and the recognition process is affected by the user's dialect, accent, and recognition environment, which may cause errors in the recognized semantic recognition text, thus As a result, the query or dialogue based on speech recognition text is wrong, so it is urgent to provide a speech recognition error correction scheme

Images