A processing system and method for accurately completing the textualization of continuous natural speech

Abstract

The invention belongs to the technical field of speech textualization, and in particular relates to a processing system and method for accurately completing the textualization of continuous natural speech. The processing system includes a cloud speech recognition engine and a post speech recognition modification platform, and the post speech recognition modification platform is connected to the cloud speech recognition engine. The invention has the following effects: real-time collection of audio, video and voice information, information collection is uninterrupted, and at the millisecond level, the collection rate reaches 100%; the information loss rate is 0; Establish a scientific and convenient human-computer interaction system for speech and text; achieve 100% conversion rate of speech and text and an accuracy rate of over 99.7%; create three-dimensional electronic integrated documents of audio, video and text.

Description

technical field [0001] The invention belongs to the technical field of textualization of speech, and in particular relates to a processing system and method for accurately completing textualization of continuous natural speech. Background technique [0002] With the continuous development of information processing technology, human-computer interaction in natural language has become a reality. The key to realizing human-computer interaction is to accurately understand the natural language instructions issued or obtained by users and perform corresponding operations. After the user issues or obtains a natural language instruction, the instruction is converted into a voice text. For more than 100 years in our country, how to convert continuous natural speech into text shorthand in real time has been a problem that people continue to explore and study. [0003] The main carrier of the current shorthand is the computer. However, after years of practice, the computer shorthand ...

AI Technical Summary

Problems solved by technology

[0012] (2) There are many local dialects of Chinese phonetics

[0015] (3) The level of Mandarin varies from person to person, and the correct rate of speech recognition also varies from person to person, not 100% accurate

Background noise accompanied by human voice, physical noise of transmission equipment, too strong or too weak voice input volume will all affect the accuracy of speech recognition

Images